Ga32 0450 09

IBM TotalStorage LTO Ultrium
Tape Drive
SCSI Reference
GA32-0450-09
IBM TotalStorage LTO Ultrium
Tape Drive
SCSI Reference
GA32-0450-09
Note
Before using this manual and the product it supports, read the information under Appendix B, “Notices,” on page 307.
Tenth Edition (May 2008)

This edition applies to the IBM TotalStorage LTO Ultrium Tape Drive SCSI Reference and to all subsequent releases and
modifications unless otherwise indicated in new editions.
© Copyright International Business Machines Corporation 2002, 2008. All rights reserved.
US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract
with IBM Corp.
Read This First
This is the tenth edition of the IBM TotalStorage LTO Ultrium Tape Drive SCSI
Reference (May 2008).
Summary of Changes
Second Edition, April 2004

The following changes were made in this edition:
v Added features of the 3580 Tape Drive Models L23 and H23.
v Modified to show that the generation 2 drive code ignores the Page Control (PC)
field.
v Modified parameters and descriptions in Table 53 on page 58.
v Modified Table 54 on page 59 to show added support for TapeAlerts 51, 52, and
53 and to show removed claim support for flags 19, 35, and 38.
v Modified the Medium Type field definition in section “Mode Parameter Header”
on page 110.
v Modified sections“Mode Page 0Fh: Data Compression Mode Page” on page 118
and “Mode Page 10h: Sequential Access Device Configuration Page” on page 120
to reflect that now the Emerald drive’s behavior matches that of the Sapphire
drive.
v Modified the Text Generation value in Table 150 on page 164.
v Modified section Table 151 on page 165 to show that Medium Auxiliary Memory
parameters 1400h-17FFh of Application Specific Data can be written to Cartridge
Memory as long as space is available.
v Modified section“READ POSITION” on page 180 to correctly state the behavior
of the READ POSITION command after a READ type command.
v Modified section “SET CAPACITY (not Ultrium 1)” on page 248 to show that
Check Condition status is returned with Illegal Field in CDB (5/2400h) if the
Capacity Proportion Value of a new cartridge becomes smaller than 17.2 GB.
Third Edition, December 2004

v Added Generation 3 (Ultrium 3 Support).
v Added LTO Gen 3 log pages, as appropriate.
v Added support for IBM® TotalStorage® 3580 Tape Drive Models L33/L3H.
v Added support for IBM TotalStorage Ultrium 3 Tape Drive.
v Added support for IBM TotalStorage 400® GB Data Cartridge.
v Added new required items to Inquiry pages in the section entitled “INQUIRY”
on page 21.
v Changed values of BQue field and CmdQ field in the section entitled “Standard
Inquiry Data” on page 22.
v Added new log sense data and additional supported log pages (see “LOG
SENSE” on page 50).
v Replaced TapeAlert Log Parameters in Table 54 on page 59.
© Copyright IBM Corp. 2002, 2008 iii

v Created a mode page that can be saved and used to configure behavior changes
to a drive (see “Mode Page 2Fh: Behavior Configuration Mode Page” on page
134).
v Added speed matching and defined a byte field to be used to distinguish
between a full performance drive and a non-performance limited drive (see
“Sense Data Format” on page 194).
Fourth Edition, March 2005

v Added Fibre Channel support.
Fifth Edition, May 2006

v Added Space (16) command to Chapter 3.
v Added Chapter 8 Firmware Download.
Sixth Edition, December 2006

v Added additional sense values to Chapter 7, “Sense Keys and Additional Sense,”
on page 269
v Added support for the Tape Data Encryption Security Protocol of the Security
Protocol In and Security Protocol Out commands
Seventh Edition, June 2007

Information to update this edition was obtained from the following Functional
Change Request documents (FCRs):
v FCR 3056 - LTO Emulate WORM Additions
v FCR 3048 - LTO WORM Read Attribute Behavior
v FCR 3059r1 - Add support for Load_Unload Hold bit
v FCR 3078 - SPC-4 Set Timestamp and Report Timestamp
v FCR 3080 - Encryption Counters
v FCR 3089 - Component Revision Level Inquiry Page
v FCR 3096 - LTO TapeAlert 22 Clearing Behavior Change
v FCR 3097 - LTO Limit ERP
v FCR 3099 - LTO HH Changes
v FCR 3103 - New Thread_Unthread Additional Sense Code
v FCR 3104 - Log Pages 14h & 16h
v FCR 3105 - LTO Set Capacity - Return to LP4
v FCR 3106 - Log Page 11h on LUN 0
v FCR 3107 - Allow Read Attribute when CM Accessible
v FCR 3108 - LTO Read dump with strictly increasing offsets
v FCR 3109r1 - Send Diagnostic (Read Thermal Sensor)
v FCR 3112 - LTO4 (J2E Encryption Method)
v FCR 3121r2 - License Keys (LTO) (Internal with Inquiry Data)
v FCR 3122 - LTO Log Page 30h - Tape Usage Log Updates
v FCR 3123 - Component Revision Level Inquiry Page
v FCR 3124 - LTO Dead Media reporting (page 2F)
iv IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

v FCR 3128 - LTO Log Page 31 Clarification
v FCR 3130 - LTO pSCSI U320 Support
v FCR 3132 - Mode Parameter Speed Field
Eighth Edition, July 2007

Information to update this edition was obtained from the following Functional
Change Request documents (FCRs):
v FCR 3133 - LTO Inquiry page C8h
v FCR 3136r0 - LTO ITD Pages (Entire Set)
v FCR 3112 - LTO 4 (J2E Encryption Method)
v FCR 3130 - LTO pSCSI U320 Support
v FCR 3131 - Direct Key Labels
v FCR 3139 - Mode Page 2F Cleanup for Pubs.
Ninth Edition, August/September 2007

Changes and additions to the following Chapters and subsections in the ninth
edition are listed as follows:
v Related Publications section
– The last bulleted item is changed to reference the Ultrium 4 tape drive.
v Chapter 1. Introduction:
– The second paragraph now includes Serial Attached SCSI (SAS) Interface.
– The subsection titled SCSI and Fibre Channel Attachment was changed to
Primary Interface Attachment.
v Chapter 2. Summary of Drive Generation Differences
– Added WORM capability and Fibre Channel information
– Updated Ultrium Tape Drive features
– Added information to Table 4. Command Timeout Values (Ultrium 1, 2 and 3
Full-Height) - Sorted by Circle
v Chapter 3. Command Support
– Standard Inquiry Data subsection: modified product revision level
information.
– Supported Vital Product Data Page subsection: additional information was
added to the Bit next to Byte 9 in the Supported Vital Product Data Inquiry
Page table.
– Firmware Design subsection: corrected the format for cross-references.
– Device Identification subsection: removed markers for external document
version.
– Mode Page 2 Fh, Behavior configuration Mode Page: added information from
FCR 3139.
– SET CAPACITY (not Ultrium 1) subsection: modified Minimum Capacity
Proportion Values.
v Chapter 4. Error Sense Information
– Fencing Behavior subsection: modified to reflect updated Ultrium version
information
v Chapter 9. Attachment Features
– Features of the Fibre Channel Interface subsection: added updated Ultrium
information
v FCR 3140 - LTO Crypto Diagnostics.
Read This First v

v Updated Chapter 7 - Sense Keys and Additional Sense
Tenth Edition, May 2008

Changes and additions to the following Chapters and subsections in the tenth
edition are listed as follows:
v FCR 3157: eServer information to be externalized
vi IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Contents
Read This First . . . . . . . . . . . iii Device Identification Page (83h) . . . . . . 39
Summary of Changes . . . . . . . . iii . . . Drive Component Revision Levels Pages (C0h) 44
Second Edition, April 2004 . . . . . iii . . . LOAD/UNLOAD . . . . . . . . . . . . 45
Third Edition, December 2004 . . . . iii . . . LOCATE . . . . . . . . . . . . . . . 47
Fourth Edition, March 2005 . . . . . iv . . . LOG SELECT . . . . . . . . . . . . . . 49
Fifth Edition, May 2006 . . . . . . iv . . . LOG SENSE . . . . . . . . . . . . . . 50
Sixth Edition, December 2006 . . . . iv . . . Log Page Format . . . . . . . . . . . 51
Seventh Edition, June 2007 . . . . . iv . . . Log Page 00h: Supported Log Pages . . . . . 52
Eighth Edition, July 2007 . . . . . . . . . v Log Page 02h: Write Error Counters . . . . . 55
Ninth Edition, August/September 2007 . . . . v Log Page 03h: Read Error Counters . . . . . 56
Tenth Edition, May 2008 . . . . . . . . . vi Log Page 06h: Non-Medium Errors (Not Ultrium
1 or Ultrium 2 non-RoHS) . . . . . . . . 57
Tables . . . . . . . . . . . . . . . xi Log Page 0Ch: Sequential Access Device Log . . 58
Log Page 2Eh: Tape Alert . . . . . . . . . 59
Log Page 11h: DT Device Status Log Page . . . 62
Preface . . . . . . . . . . . . . . xv Log Page 14h: Device Statistics Log Page (Not
Organization . . . . . . . . . . . . . . xv Ultrium 1 or 2) . . . . . . . . . . . . 69
Related Publications . . . . . . . . . . . xv Log Page 16h: Tape Diagnostic Data Log Page
(Not Ultrium 1 or 2) . . . . . . . . . . 71
Chapter 1. Introduction . . . . . . . . 1 Log Page 18h: Protocol-Specific Logical Unit Log
Supported Servers and Operating Systems . . . . 3 Page . . . . . . . . . . . . . . . . 74
Primary Interface Attachment . . . . . . . . 3 Log Page 30h: Tape Usage Log . . . . . . . 77
Supported Device Drivers . . . . . . . . . . 4 Log Page 31h: Tape Capacity Log . . . . . . 79
Supported Tape Cartridges . . . . . . . . . 5 Log Page 32h: Data Compression Log . . . . 80
Log Page 33h: Write Errors (Not Ultrium 1 or
Chapter 2. Summary of Drive Generation Ultrium 2 non-RoHS) . . . . . . . . . . 81
Differences . . . . . . . . . . . . . 7 Log Page 34h: Read Forward Errors (Not Ultrium
1 or Ultrium 2 non-RoHS) . . . . . . . . 83
Differences in Command Timeout Values . . . . . 8
Log Page 37h: Performance Characteristics
Command and Parameter Differences Between
(Ultrium 4) . . . . . . . . . . . . . 85
Generations . . . . . . . . . . . . . . 15
Log Page 38h: Blocks/Bytes Transferred (Not
Data Changes Between Generations . . . . . . 15
Ultrium 1 or Ultrium 2 non-RoHS) . . . . . 95
Standard Inquiry Data . . . . . . . . . . 15
Log Page 39h: Host Port 0 Interface Errors (Not
REPORT DENSITY SUPPORT Command . . . 15
Mode Pages . . . . . . . . . . . . . 16
READ POSITION Command . . . . . . . 16
+ Log Page 3Ah: Drive Control Verification (Not in
Request Sense Changes . . . . . . . . . 16
+ Ultrium 1) . . . . . . . . . . . . . . 98
Log Page 3Bh: Host Port 1 Interface Errors (Not
Behavior Changes Between Generations . . . . . 16
Cartridge Eject for Errors . . . . . . . . . 16
Log Page 3Ch: Drive Usage Information . . . 100
Queueing Issues . . . . . . . . . . . . 16
Log Page 3Dh: Subsystem Statistics (Not
Microcode Detection of Errors . . . . . . . . 16
Fencing Behavior . . . . . . . . . . . 16
+ Log Page 3Eh: Engineering Use (Not Ultrium 1
+ or Ultrium 2 non-RoHS) . . . . . . . . . 104
Chapter 3. Command Support . . . . . 17 Log Page 3Eh[3Ch]: Drive Control Statistics
ERASE . . . . . . . . . . . . . . . . 20 (Not Ultrium 1 or Ultrium 2 non-RoHS) . . . 105
INQUIRY . . . . . . . . . . . . . . . 21 MODE SELECT . . . . . . . . . . . . 106
Standard Inquiry Data . . . . . . . . . . 22 MODE SENSE . . . . . . . . . . . . . 107
Supported Vital Product Data Page . . . . . 27 Mode Data Format . . . . . . . . . . 109
Firmware Designation Page (03h) . . . . . . 28 Mode Parameter Header . . . . . . . . . 110
Mode Page Policy Page (87h) . . . . . . . 29 Mode Block Descriptor . . . . . . . . . 112
SCSI Ports Page (88h) . . . . . . . . . . 32 Mode Page 01h: Read-Write Error Recovery
Sequential-Access Device Capabilities Page (B0h) 34 Page . . . . . . . . . . . . . . . 113
Drive Serial Numbers (C1h) . . . . . . . . 35 Mode Page 02h: Disconnect/Reconnect Page . . 115
Device Unique Configuration Data Page (C7h) . 36 Mode Page 0Ah: Control Extension Mode Page 117
Mode Parameter Default Settings Page (C8h) . . 37 Mode Page 0Fh: Data Compression Mode Page 118
Unit Serial Number Page (80h) . . . . . . . 38
© Copyright IBM Corp. 2002, 2008 vii

Mode Page 10h: Sequential Access Device MIM Data Structure . . . . . . . . . . 227
Configuration Page . . . . . . . . . . 120 Self-Test . . . . . . . . . . . . . . 230
Mode Page 18h: Protocol-Specific Logical Unit Post A Self Test Diagnostic . . . . . . . . 231
Control Page . . . . . . . . . . . . 123 Post B Performance Diagnostic . . . . . . 232
Mode Page 19h: Protocol Specific Port Control Post C Media Test Diagnostic . . . . . . . 234
Page . . . . . . . . . . . . . . . 124 Post D Head Test Diagnostic . . . . . . . 235
Mode Page 19h: Protocol Specific Port Control Primary Port Wrap Test Diagnostic . . . . . 236
Page Short Format (Ultra 320 Attached Drives Force Dump . . . . . . . . . . . . . 237
Only) . . . . . . . . . . . . . . . 126 Write Dump To Cartridge . . . . . . . . 238
Mode Page 19h[03h]: Negotiated Settings Mode Write Dump to Flash (EEPROM) . . . . . . 240
Subpage . . . . . . . . . . . . . . 127 Clear FLASH Dump (EEPROM) . . . . . . 241
Mode Page 19h[04h]: Report Transfer Set Traps . . . . . . . . . . . . . . 242
Capabilities Mode Subpage . . . . . . . . 129 Remove Traps . . . . . . . . . . . . 243
Mode Page 1Ch: Information Exceptions Mode Reset Drive . . . . . . . . . . . . . 245
Page . . . . . . . . . . . . . . . 130 SET CAPACITY (not Ultrium 1) . . . . . . . 248
Mode Page 1Dh: Medium Configuration Mode SET TIMESTAMP (Not LTO 1). . . . . . . . 250
Page . . . . . . . . . . . . . . . 131 SPACE . . . . . . . . . . . . . . . 252
Mode Page 24h: Vendor-Specific Control Mode TEST UNIT READY . . . . . . . . . . . 254
Page . . . . . . . . . . . . . . . 133 VERIFY . . . . . . . . . . . . . . . 255
Mode Page 2Fh: Behavior Configuration Mode WRITE . . . . . . . . . . . . . . . 256
Page . . . . . . . . . . . . . . . 134 WRITE ATTRIBUTE . . . . . . . . . . . 257
Mode Page 3Dh: LEOT Mode Page (LTO 1 WRITE BUFFER . . . . . . . . . . . . 258
Only) . . . . . . . . . . . . . . . 143 WRITE FILE MARKS . . . . . . . . . . . 260
PERSISTENT RESERVE IN . . . . . . . . . 144
PERSISTENT RESERVE OUT . . . . . . . . 149 Chapter 4. Error Sense Information 261
PREVENT/ALLOW MEDIUM REMOVAL . . . 153 Sense Data . . . . . . . . . . . . . . 261
READ . . . . . . . . . . . . . . . . 154 Sense Data Management . . . . . . . . . 261
READ ATTRIBUTE . . . . . . . . . . . 156 Unit Attention Conditions . . . . . . . . . 261
Vendor-Specific Medium Type Attributes . . . 158 Persistent Errors . . . . . . . . . . . . 262
Format for the Attribute Values Service Action 159 Fencing Behavior . . . . . . . . . . . 262
Format for the Attribute List Service Action . . 160
Format for the Volume List Service Action . . . 161
Chapter 5. Implementation
Format for the Partition List Service Action . . 162
Medium Auxiliary Memory . . . . . . . 163 Considerations . . . . . . . . . . . 265
READ BLOCK LIMITS . . . . . . . . . . 172 Device Clocks . . . . . . . . . . . . . 265
READ BUFFER . . . . . . . . . . . . . 173
Error Log Buffer (06h) . . . . . . . . . 177 Chapter 6. WORM BEHAVIORS . . . . 267
World Wide Name Buffer (08h) . . . . . . 178 Conditions for Writing . . . . . . . . . . 267
SCSI Log Buffer (07h) . . . . . . . . . 179 Command Behavior When WORM Medium Has
READ POSITION . . . . . . . . . . . . 180 Been Tampered With . . . . . . . . . . . 267
RECEIVE DIAGNOSTIC RESULTS . . . . . . 183
RELEASE UNIT . . . . . . . . . . . . 184 Chapter 7. Sense Keys and Additional
REPORT DENSITY SUPPORT . . . . . . . . 185 Sense. . . . . . . . . . . . . . . 269
REPORT LUNs . . . . . . . . . . . . . 190
Sense Key 0 (No Sense) . . . . . . . . . . 269
REPORT SUPPORTED TASK MANAGEMENT
Sense Key 1 (Recovered Error) . . . . . . . . 270
FUNCTIONS (Not LTO1) . . . . . . . . . 192
Sense Key 2 (Not Ready) . . . . . . . . . 271
REPORT TIMESTAMP (Not LOT 1) . . . . . . 193
Sense Key 3 (Medium Error) . . . . . . . . 271
REQUEST SENSE . . . . . . . . . . . . 194
Sense Key 4 (Hardware or Firmware Error) . . . 273
Sense Data Format . . . . . . . . . . 194
Sense Key 5 (Illegal Request) . . . . . . . . 273
RESERVE UNIT . . . . . . . . . . . . 199
Sense Key 6 (Unit Attention) . . . . . . . . 275
REWIND . . . . . . . . . . . . . . . 200
Sense Key 7 (Data Protect) . . . . . . . . . 275
SECURITY PROTOCOL IN (SPIN) A2h . . . . . 201
Sense Key 8 (Blank Check) . . . . . . . . . 276
Security Protocol In Pages (00h - Security
Sense Key B (Aborted Command) . . . . . . 277
Protocol Information) . . . . . . . . . . 202
Sense Key D (Volume Overflow) . . . . . . . 277
SECURITY PROTOCOL OUT (SPOUT) B5h . . . 215
Security Protocol Out Pages (20h - Tape Data
Encryption Security Protocol) . . . . . . . 216 Chapter 8. Device Hardware
SPOUT (20h:0010h) - Set Data Encryption . . . 216 Encryption (Ultrium 4) . . . . . . . 279
Key-Associated Data (KAD) Descriptors . . . 219
SEND DIAGNOSTIC . . . . . . . . . . . 223 Chapter 9. Attachment Features . . . 281
SIM Data Structure . . . . . . . . . . 224 Types of Interface Attachments . . . . . . . 281
viii IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Common Tape LUN Behaviors . . . . . . . 281 ASN.1 [DER] Encoding Implementation . . . . 297
Power-On . . . . . . . . . . . . . 281 SEDK ASN.1 Format . . . . . . . . . . . 297
Reset Strategy . . . . . . . . . . . . 281 ASN.1 Standalone SEDK (SEDK Only, Clear,
Abort Handling . . . . . . . . . . . 282 Signed) - Example . . . . . . . . . . . . 298
Multi-initiator Support . . . . . . . . . 284 ASN.1 Standalone SEDK (SEDK Only, dSK
Status Codes . . . . . . . . . . . . 284 Encrypted, Signed) - Example . . . . . . . . 299
Features of the SCSI Interface . . . . . . . . 285 xAK/xSK ASN.1 Format . . . . . . . . . 300
LUN Identification . . . . . . . . . . 285 ASN.1 xAK/xSK Structure (Clear, Signed) -
SPI Information Unit (Ultra 320 Attached Drives Example . . . . . . . . . . . . . . . 301
Only) . . . . . . . . . . . . . . . 285 Signing ASN.1 Format . . . . . . . . . . 302
Bus Parity Errors (When Using Data Group ASN.1 Signing Information (Clear, Unsigned) -
Transfers). . . . . . . . . . . . . . 285 Example . . . . . . . . . . . . . . . 303
Bus Parity Errors (When Using Information Unit ASN.1 Signing Information (eAK Encrypted, dAK
Transfers). . . . . . . . . . . . . . 286 Signed) - Example . . . . . . . . . . . . 303
Disconnect Strategy . . . . . . . . . . 286 HMAC ASN.1 Format . . . . . . . . . . 304
Messages . . . . . . . . . . . . . . 287 ASN.1 HMAC Structure - Example . . . . . . 304
Features of the Fibre Channel Interface . . . . . 291
Features of the Serial Attached SCSI (SAS) Interface 291 Appendix B. Notices . . . . . . . . 307
Trademarks . . . . . . . . . . . . . . 307
Chapter 10. Firmware Download . . . 293
Identifying Level Hardware of Drive (Not Ultrium Glossary . . . . . . . . . . . . . 309
1) . . . . . . . . . . . . . . . . . 293
Identifying the Level Hardware for Which the
Index . . . . . . . . . . . . . . . 317
Firmware Image is Intended (Not Ultrium 1) . . . 295
Download Process . . . . . . . . . . . 295
Appendix A. ASN.1 Information (Used

for Encryption) . . . . . . . . . . . 297
Contents ix
x IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference
Tables
1. Features of the IBM Ultrium Tape Drives and 47. Supported Log Pages Log Page Format
the IBM 3580 Ultrium Tape Drive . . . . . 2 (Ultrium 1 and Ultrium 2 non-RoHS) . . . . 52
2. Supported Servers and Operating Systems for 48. Supported Log Pages Log Page Format
Primary Interface Attachment . . . . . . . 3 (Ultrium 2 RoHS and Ultrium 3) . . . . . 52
3. IBM LTO Ultrium Tape Drive Support of LTO 49. Supported Log Pages Log Page Format
Cartridges . . . . . . . . . . . . . 5 (Ultrium 4) . . . . . . . . . . . . . 53
4. Command Timeout Values (Ultrium 1, 2, and 3 50. Write Error Log Parameters . . . . . . . 55
Full-Height) - Sorted by OpCode . . . . . . 8 51. Read Error Log Parameters . . . . . . . 56
5. Command Timeout Values (Ultrium 3 52. Non-Medium Errors log parameter codes 57
Half-Height and Ultrium 4) - Sorted by 53. Sequential Access Device Log Parameters 58
OpCode . . . . . . . . . . . . . . 10 54. TapeAlert Log Parameters . . . . . . . . 59
6. Command Timeout Values (Ultrium 3 55. DT Device Status Log Page . . . . . . . 62
Half-Height and Ultrium 4) - Alphabetical . . 12 56. DT Device Status log page parameter codes 62
7. Command and Parameter differences between 57. Very high frequency data log parameter format 62
generations . . . . . . . . . . . . 15 58. VHF DATA DESCRIPTOR field . . . . . . 63
8. Supported Common SCSI Commands . . . . 17 59. DT DEVICE ACTIVITY field values . . . . 65
9. ERASE Command . . . . . . . . . . 20 60. Very high frequency polling delay log
10. INQUIRY Command . . . . . . . . . 21 parameter format . . . . . . . . . . 66
11. Standard Inquiry Data Valid LUN . . . . . 22 61. Primary Port Status descriptor . . . . . . 66
12. Standard Inquiry Data Valid LUN (Drive 62. Speed Values . . . . . . . . . . . . 68
configured with eServer attachment feature) . 24 63. Device Statistics log parameter codes . . . . 69
13. Product Identification Table . . . . . . . 26 64. Medium Type Log Parameter format . . . . 69
14. Supported Vital Product Data Inquiry Page 27 65. Medium Type log parameter . . . . . . . 70
15. Firmware Designation Page . . . . . . . 28 66. Tape Diagnostic Data log page . . . . . . 71
16. Mode Page Policy page . . . . . . . . 29 67. Tape Diagnostic Data log parameter format 72
17. Mode Page policy descriptor . . . . . . . 29 68. Protocol-Specific log page for SAS . . . . . 74
18. Mode page policy values . . . . . . . . 30 69. Protocol-Specific log parameter for SAS 75
19. Mode Page Policy for LUN 0 (SSC Device) 30 70. SAS Phy log descriptor. . . . . . . . . 76
20. Mode Page Policy LUN 2 (ADC Device) 30 71. Tape Usage Log Parameters . . . . . . . 77
21. SCSI Ports Inquiry Page . . . . . . . . 32 72. Tape Capacity Log Parameters . . . . . . 79
22. SCSI Port Identification Descriptor . . . . . 33 73. Data Compression Log Parameters . . . . . 80
23. Target Port Descriptor Format . . . . . . 33 74. Write Errors log parameter codes . . . . . 81
24. Sequential-Access Device Capabilities Page 34 75. Read Forward Error Counters log parameter
25. Inquiry Page C1h: Drive Serial Numbers 35 codes . . . . . . . . . . . . . . 83
26. Unit Serial Number Inquiry Page . . . . . 38 76. Log Page 37h: Performance Characteristics:
27. Device Identification Inquiry Page . . . . . 39 Quality Summary . . . . . . . . . . 87
28. Device Identification Descriptor Format 39 77. Log Page 37h: Performance Characteristics:
29. Identifier Format . . . . . . . . . . . 40 Device Usage . . . . . . . . . . . . 87
30. Protocol Identification descriptor . . . . . 40 78. Log Page 37h: Performance Characteristics:
31. Logical Unit (NAA) Descriptor Format 41 Host Commands . . . . . . . . . . . 88
32. Port Identifier Descriptor Format . . . . . 41 79. Log Page 37h: Performance Characteristics:
33. Relative Target Port Values . . . . . . . 41 Host Initiators . . . . . . . . . . . 91
34. Port Name (NAA) Descriptor Format . . . . 42 80. Log Page 37h: Performance Characteristics:
35. Target Device Name Descriptor . . . . . . 43 Host Recovery (by port) . . . . . . . . 91
36. Drive Component Revision Levels Pages 44 81. Log Page 37h: Performance Characteristics:
37. LOAD/UNLOAD Command . . . . . . 45 Mode Phase Timing Windows . . . . . . 91
38. LOAD/UNLOAD actions depending on Load 82. Log Page 37h: Performance Characteristics:
and Hold bit settings and medium position . . 46 Servo Speed Characteristics . . . . . . . 94
39. LOCATE (10) command . . . . . . . . 47 83. Log Page 37h: Performance Characteristics:
40. LOCATE (16) Command (Ultrium 4 and Later) 47 Static Capacity . . . . . . . . . . . 94
41. Dest_Type field definitions . . . . . . . 48 84. Log Page 37h: Performance Characteristics:
42. LOG SELECT Command . . . . . . . . 49 Active Capacity . . . . . . . . . . . 94
43. LOG SENSE Command . . . . . . . . 50 85. Blocks/Bytes Transferred log parameter codes 95
44. Log Page Header Format . . . . . . . . 51 86. Host Port Interface log parameter codes 97
45. Log Parameter Format . . . . . . . . . 51 87. Subsystem Statistics log parameter codes 101
46. Supported Log Pages Log Page Definition 52 88. 6-Byte MODE SELECT Command . . . . 106
© Copyright IBM Corp. 2002, 2008 xi

89. 10-Byte MODE SELECT Command . . . . 106 137. PERSISTENT RESERVE OUT command with
90. 6-Byte MODE SENSE Command . . . . . 107 REGISTER AND MOVE service action
91. 10-Byte MODE SENSE Command . . . . . 107 parameter list . . . . . . . . . . . 152
92. Mode Data Format . . . . . . . . . . 109 138. PREVENT/ALLOW MEDIUM REMOVAL
93. Mode Parameter Header 6-Byte Command 110 Command . . . . . . . . . . . . 153
94. Mode Parameter Header 10-Byte Command 110 139. READ Command . . . . . . . . . . 154
95. Media Type Values . . . . . . . . . . 110 140. READ ATTRIBUTE Command . . . . . . 156
96. Mode Block Descriptor . . . . . . . . 112 141. Supported Service Action Codes . . . . . 157
97. Read-Write Error Recovery Page . . . . . 113 142. Vendor-Specific Medium Type Attributes 158
98. Read Retry Limit Definition . . . . . . . 113 143. Parameter Data for an Attribute Values
99. Write Retry Limit Definition . . . . . . 114 Service Action . . . . . . . . . . . 159
100. Disconnect/Reconnect Mode Page . . . . 115 144. Parameter Data for an Attribute List Service
101. Control Extension Mode Page . . . . . . 117 Action . . . . . . . . . . . . . . 160
102. Data Compression Mode Page . . . . . . 118 145. Parameter Data for a Volume List Service
103. Responses to Data Boundaries . . . . . . 119 Action . . . . . . . . . . . . . . 161
104. Sequential Access Device Configuration Page 120 146. Parameter Data for a Partition List Service
105. Fibre Channel Logical Unit Control Page 123 Action . . . . . . . . . . . . . . 162
106. SAS Logical Unit Control Page . . . . . . 123 147. Types of MAM Attributes . . . . . . . 163
107. Fibre Channel Port Control Page . . . . . 124 148. States for the Types of MAM Attributes 163
108. SAS Port Control Page . . . . . . . . 125 149. Format of a MAM Attribute. . . . . . . 164
109. Protocol Specific Port Control Page short 150. Values and Requirements for the Format Field 164
format (U320 only) . . . . . . . . . . 126 151. Range of Values for MAM Attribute
110. Negotiated Settings mode subpage (U320 Identifiers . . . . . . . . . . . . . 165
drives only) . . . . . . . . . . . . 127 152. Standard Device Type Attributes . . . . . 166
111. Transfer Period Factor . . . . . . . . 128 153. Format for Device Vendor
112. Transfer Width Exponent. . . . . . . . 128 Identification/Product Serial Number
113. Report Transfer Capabilities mode subpage Attribute, Device Vendor
(U320 drives only) . . . . . . . . . . 129 Identification/Product Serial Number at
114. Information Exceptions Mode Page . . . . 130 Load-1 Attribute, Device Vendor
115. Medium Configuration Mode Page . . . . 131 Identification/Product Serial Number at
116. Worm Mode Label Restrictions field values 131 Load-2 Attribute, and Device Vendor
117. Worm Mode Filemarks Restrictions field Identification/Product Serial Number at
values . . . . . . . . . . . . . . 132 Load-3 Attribute . . . . . . . . . . 168
118. Vendor-Specific Control Mode Page . . . . 133 154. Standard Medium Type Attributes . . . . 168
119. Behavior Configuration Mode Page . . . . 134 155. Values for Medium Type and Medium Type
120. Fence Behavior Selection Values . . . . . 135 Information Attributes . . . . . . . . 169
121. Periodic Clean Notification Usage Criteria 137 156. Standard Host Type Attributes . . . . . . 169
122. WORM Behavior . . . . . . . . . . 138 157. Values for the Text Localization Identifier
123. Sense Data Behavior . . . . . . . . . 139 Attribute . . . . . . . . . . . . . 170
124. Unload On Error Behavior Definition 142 158. Vendor-Specific Medium Type Attributes 171
125. LEOT Mode Page (Ultrium 1 Only) . . . . 143 159. READ BLOCK LIMITS Command . . . . 172
126. PERSISTENT RESERVE IN Command 144 160. READ BLOCK LIMITS Descriptor . . . . 172
127. PERSISTENT RESERVE IN Parameter Data 161. READ BUFFER Command . . . . . . . 173
for Read Capabilities . . . . . . . . . 144 162. Description of the Mode Field . . . . . . 173
128. PERSISTENT RESERVE IN Parameter Data 163. READ BUFFER Header . . . . . . . . 174
for Read Full Status . . . . . . . . . 145 164. READ BUFFER Descriptor . . . . . . . 174
129. PERSISTENT RESERVE IN Full Status 165. READ ECHO BUFFER Descriptor . . . . . 175
descriptor format . . . . . . . . . . 146 166. Drive Buffers . . . . . . . . . . . 176
130. PERSISTENT RESERVE IN Parameter Data 167. Error Log Buffer Command . . . . . . . 177
for Read Keys . . . . . . . . . . . 146 168. World Wide Name Buffer . . . . . . . 178
131. PERSISTENT RESERVE IN Parameter Data 169. Error Log Buffer Command . . . . . . . 179
for Read Reservations . . . . . . . . . 147 170. READ POSITION Command . . . . . . 180
132. PERSISTENT RESERVE IN Read Reservations 171. READ POSITION Data . . . . . . . . 181
Descriptor . . . . . . . . . . . . 147 172. RECEIVE DIAGNOSTIC RESULTS Command 183
133. PERSISTENT RESERVE OUT Command 149 173. 6-Byte RELEASE UNIT Command . . . . 184
134. Values for Service Action Codes in 174. 10-Byte RELEASE UNIT Command . . . . 184
PERSISTENT RESERVE OUT Command . . 149 175. REPORT DENSITY SUPPORT Command 185
135. Persistent Reservation Type Codes . . . . 150 176. REPORT DENSITY SUPPORT Header 186
136. PERSISTENT RESERVE OUT Parameter List 151 177. REPORT DENSITY SUPPORT Descriptor
Block . . . . . . . . . . . . . . 186
178. Density Information for LTO Formats 187
xii IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

179. REPORT LUNs Command . . . . . . . 190 226. Post C Media Test Send Diagnostic Parameter
180. Logical Unit Numbers Data . . . . . . . 190 Data . . . . . . . . . . . . . . 234
181. REPORT SUPPORTED TASK 227. Post C Media Test Receive Diagnostic
MANAGEMENT FUNCTIONS command . . 192 Parameter Data . . . . . . . . . . . 234
182. REPORT SUPPORTED TASK 228. Post D Head Test Send Diagnostic Parameter
MANAGEMENT FUNCTIONS parameter Data . . . . . . . . . . . . . . 235
data . . . . . . . . . . . . . . . 192 229. Post D Head Test Receive Diagnostic
183. REPORT TIMESTAMP command . . . . . 193 Parameter Data . . . . . . . . . . . 235
184. Timestamp Descriptor. . . . . . . . . 193 230. Primary Port Wrap Test Send Diagnostic
185. REQUEST SENSE Command . . . . . . 194 Parameter Data . . . . . . . . . . . 236
186. Sense Data Format . . . . . . . . . . 194 231. Primary Port Wrap Test Receive Diagnostic
187. Volume Label Cartridge Type . . . . . . 198 Parameter Data . . . . . . . . . . . 236
188. 6-Byte RESERVE UNIT Command . . . . 199 232. Force Dump Diagnostic Parameter Data 237
189. 10-Byte RESERVE UNIT Command . . . . 199 233. Write Dump to Cartridge Send Diagnostic
190. REWIND Command . . . . . . . . . 200 Parameter Data . . . . . . . . . . . 238
191. Security Protocol In - A2h CDB . . . . . 201 234. Write Dump to Cartridge Receive Diagnostic
192. Security Protocol Definitions . . . . . . 201 Parameter Data . . . . . . . . . . . 239
193. Security Protocol Specific Definitions for 235. Write Dump To FLASH Send Diagnostic
Security Protocol 00h . . . . . . . . . 202 Parameter Data . . . . . . . . . . . 240
194. Supported Security Protocols List Structure 203 236. Write Dump To FLASH Receive . . . . . 240
195. Certificate Data Structure . . . . . . . 204 237. Clear Dump to FLASH Send Diagnostic
196. Security Protocol Specific Definitions for Parameter Data . . . . . . . . . . . 241
Security Protocol 20h . . . . . . . . . 205 238. Set Traps Diagnostic Parameter Data 242
197. 0000h - Tape Data Encryption In Support 239. Remove Traps Diagnostic Parameter Data 243
Pages Structure . . . . . . . . . . . 206 240. Reset Drive Diagnostic Results Data . . . . 244
198. Tape Data Encryption In page codes . . . . 206 241. Reset Drive Diagnostic Parameter Data 245
199. 0001h - Tape Data Encryption Out Support 242. Read Thermal Sensor Diagnostic parameter
Pages Structure . . . . . . . . . . . 207 data . . . . . . . . . . . . . . . 246
200. Tape Data Encryption Out page codes 207 243. Read Thermal Sensor Receive Diagnostic
201. 0010h - Data Encryption Capabilities page 208 parameter data . . . . . . . . . . . 247
202. Data Encryption Algorithm Descriptor - 244. SET CAPACITY Command . . . . . . . 248
Standard Encryption Structure . . . . . . 209 245. Minimum Capacity Proportion Values 249
203. Supported Key Formats page Structure 210 246. SET TIMESTAMP command . . . . . . 250
204. Supported Key Formats . . . . . . . . 210 247. SET TIMESTAMP parameter list format 250
205. Plaintext Key Format Structure. . . . . . 210 248. SPACE (6) Command . . . . . . . . . 252
206. Data Encryption Management Capabilities 211 249. SPACE (16) command . . . . . . . . . 252
207. 0020h - Data Encryption Status page 212 250. TEST UNIT READY Command . . . . . 254
208. 0021h - Next Block Encryption Status page 213 251. VERIFY Command . . . . . . . . . . 255
209. SPIN (20h:0021h) - KAD Parameters by Mode 214 252. WRITE Command . . . . . . . . . . 256
210. Security Protocol Out B5h CDB . . . . . 215 253. WRITE ATTRIBUTE Command . . . . . 257
211. Security Protocol Specific Definitions for 254. Parameter Data for Attribute Values Service
Security Protocol 20h . . . . . . . . . 216 Action Request . . . . . . . . . . . 257
212. 0010h - Set Data Encryption page . . . . . 216 255. WRITE BUFFER Command . . . . . . . 258
213. SPOUT (20h:0010h) - KAD Parameters by 256. Description of Mode Field . . . . . . . 258
Mode . . . . . . . . . . . . . . 218 257. WRITE FILE MARKS Command . . . . . 260
214. KAD 00h - Unauthenticated KAD . . . . . 219 258. Fence State to Error Mapping . . . . . . 263
215. KAD 01h - DKi (Data Key Identifier) 220 259. Timestamp Origin . . . . . . . . . . 265
216. KAD 02h - Nonce . . . . . . . . . . 221 260. TIMESTAMP Format . . . . . . . . . 265
217. KAD 03h - Metadata (Raw) . . . . . . . 222 261. Behavior when the loaded medium has
218. SEND DIAGNOSTIC Command . . . . . 223 suspect integrity . . . . . . . . . . 267
219. Supported Diagnostics . . . . . . . . 223 262. ASC and ASQ Summary for Sense Key 0 (No
220. SIM Data Structure . . . . . . . . . 224 Sense) . . . . . . . . . . . . . . 269
221. MIM Data Structure . . . . . . . . . 227 263. ASC and ASQ Summary for Sense Key 1
222. Post A Self Test Send Diagnostic Parameter (Recovered Error) . . . . . . . . . . 270
Data . . . . . . . . . . . . . . 231 264. ASC and ASQ Summary for Sense Key 2 (Not
223. Post A Self Test Receive Diagnostic Parameter Ready) . . . . . . . . . . . . . . 271
Data . . . . . . . . . . . . . . 231 265. ASC and ASQ Summary for Sense Key 3
224. Post B Performance Send Diagnostic (Medium Error) . . . . . . . . . . . 271
Parameter Data . . . . . . . . . . . 232 266. ASC and ASQ Summary for Sense Key 4
225. Post B Performance Receive Diagnostic (Hardware or Firmware Error) . . . . . . 273
Parameter Data . . . . . . . . . . . 232
Tables xiii
267. ASC and ASQ Summary for Sense Key 5 272. ASC and ASQ Summary for Sense Key D
(Illegal Request). . . . . . . . . . . 273 (Volume Overflow). . . . . . . . . . 277
268. ASC and ASQ Summary for Sense Key 6 273. Abort Condition Handling . . . . . . . 282
(Unit Attention) . . . . . . . . . . . 275 274. Status Codes . . . . . . . . . . . . 284
269. ASC and ASQ Summary for Sense Key 7 275. Supported Outbound Messages . . . . . 287
(Data Protect) . . . . . . . . . . . 275 276. Supported Inbound Messages . . . . . . 289
270. ASC and ASQ Summary for Sense Key 8 277. Load ID and RU Name Designation . . . . 294
(Blank Check) . . . . . . . . . . . 276 278. Firmware Image . . . . . . . . . . 295
271. ASC and ASQ Summary for Sense Key B
(Aborted Command) . . . . . . . . . 277
xiv IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Preface
This publication contains information about how to use and program all models of
the IBM LTO Ultrium Tape Drive.
Organization
The information in this book is organized as follows:
v Chapter 1, “Introduction,” on page 1 describes the features and supported
attachments for each type of tape drive.
v Chapter 2, “Summary of Drive Generation Differences,” on page 7 lists the
differences in command timeout values between the IBM Ultrium Internal Tape
Drive, the IBM TotalStorage LTO Ultrium 2 Tape Drive, and the IBM
TotalStorage LTO Ultrium 3 Tape Drive (known respectively as the Generation 1,
Generation 2, and Generation 3 drives).
v Chapter 3, “Command Support,” on page 17 lists the SCSI commands that are
supported by the tape drives.
v Chapter 4, “Error Sense Information,” on page 261 describes the error sense
information for the tape drives.
v Chapter 7, “Sense Keys and Additional Sense,” on page 269 describes the sense
keys and additional sense information for the tape drives.
v Chapter 9, “Attachment Features,” on page 281 describes the features of the SCSI
and Fibre Channel and Serial Attached SCSI drives.
v Chapter 10, “Firmware Download,” on page 293 describes identifying level
hardware of drive and identifying level hardware for which the Firmware image
is intended.
Related Publications
v IBM TotalStorage Ultrium Tape Drive 3580 Models L23 and H23 Setup and Operator
Guide, GA32-0460, tells how to install and run the IBM 3580 Ultrium Tape Drive
Models L23 and H23.
v IBM 3580 Ultrium Tape Drive Setup, Operator, and Service Guide, GA32-0415, tells
how to install and run the IBM 3580 Ultrium Tape Drive. The guide also
describes how to administer basic service procedures.
v IBM TotalStorage LTO Ultrium 2 Tape Drive Models T400 and T400F Setup, Operator,
and Service Guide, GA32-0455, tells how to install and run the IBM Ultrium 2
Tape Drive. The guide also describes how to administer basic service procedures.
v IBM TotalStorage 3580 Tape Drive Models L33/L3H Setup, Operator, and Service
Guide, GC26-7708, tells how to install and run the IBM 3580 Tape Drive Models
L33/L3H. The guide also describes how to administer basic service procedures.
v IBM TotalStorage Ultrium 3 Tape Drive Setup, Operator, and Service Guide,
GC26-7697, tells how to install and run the IBM TotalStorage Ultrium 3 Tape
Drive. The guide also describes how to administer basic service procedures.
v IBM Ultrium Internal Tape Drive Models T200 and T200F Setup, Operator, and
Service Guide, GA32-0435, tells how to install and run the IBM Ultrium Internal
Tape Drive. The guide also describes how to administer basic service procedures.
v IBM Ultrium Device Drivers Installation and User’s Guide, GA32-0430, provides
instructions for attaching IBM-supported hardware to open-systems operating
systems. It indicates what devices and levels of operating systems are supported,
© Copyright IBM Corp. 2002, 2008 xv
gives the requirements for adapter cards, and tells how to configure servers to
use the device driver with the Ultrium family of devices.
v IBM Ultrium Device Drivers Programming Reference, GC35-0483, supplies
information to application owners who want to integrate their open-systems
applications with IBM-supported Ultrium hardware. The reference contains
information about the application programming interfaces (APIs) for each of the
various supported operating-system environments.
v Fibre Channel Arbitrated Loop (FC-AL-2), published by the American National
Standards Institute (ANSI) as NCITS 332:1999.
v Fibre Channel Tape and Tape Medium Changers (FC-TAPE), published by the
American National Standards Institute. Final draft available as T11/99-069v4 on
the web at http://www.t11.org; actual document available from ANSI as NCITS
TR-24:1999.
v Fibre Channel Protocol for SCSI, Second Version (FCP-2), published by the
American National Standards Institute and available on the web at
http://www.t10.org.
v SCSI Parallel Interface-3 (SPI-3), published by InterNational Committee on
Information Technology Standards (INCITS) and available on the web at
http://www.t10.org.
v SCSI-3 Stream Commands (SSC), published by the American National Standards
Institute and available on the web at http://www.t10.org.
v SCSI Stream Commands-2 (SSC-2), published by the American National Standards
Institute and available on the web at http://www.t10.org.
v SCSI Primary Commands-2 (SPC-2), published by the American National
Standards Institute and available on the web at http://www.t10.org.
v SCSI Primary Commands-3 (SPC-3), published by the American National
Standards Institute and available on the web at http://www.t10.org.
v Automation/Drive Interface - Commands (ADC), published by the American
National Standards Institute and available on the web at http://www.t10.org.
v Automation/Drive Interface - Commands (ADC-2), published by the American
National Standards Institute and available on the web at http://www.t10.org.
v IBM TotalStorage Ultrium 4 Tape Drive Setup, Operator, and Service Guide,
GC27-2102, tells how to install and run the IBM TotalStorage Ultrium 4 Tape
Drive. The guide also describes how to administer basic service procedures.
Portions of this manual were adapted from documentation provided by the

InterNational Committee on Information Technology Standards (INCITS).
xvi IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Chapter 1. Introduction
The products that are discussed in this book are high-performance, high-capacity
data-storage devices that connect to and provide additional storage for supported
servers. They include all models of the IBM LTO Ultrium Tape Drive.
Certain of the products use a Small Computer Systems Interface (SCSI); others use
a SCSI over Fibre Channel interface or a Serial Attached SCSI (SAS) interface.
Table 1 on page 2 lists the type of interface and other features for each product.
All products use the Small Computer Systems Interface (SCSI) Architecture Model.
The transports used are shown in Table 1 on page 2.
Figure 1 shows the IBM 3580 Ultrium Tape Drive and the IBM TotalStorage LTO
Ultrium Tape Drive Model T200. The IBM TotalStorage Ultrium Tape Drive Model
T800 looks similar to the IBM TotalStorage LTO Ultrium Tape Drive Model T200.
Figure 1. The IBM 3580 Ultrium Tape Drive and the IBM TotalStorage LTO Ultrium Tape
Drive Model T200. Model T200 resembles Model T200F, T400, and T400F in appearance. It
is shown on the right without a front bezel. The IBM TotalStorage Ultrium Tape Drive Model
T800 looks similar to the IBM TotalStorage LTO Ultrium Tape Drive Model T200.
© Copyright IBM Corp. 2002, 2008 1

Designed to perform unattended backups as well as to retrieve and archive files,
the Ultrium Tape Drives include the features that are described in Table 1.
Table 1. Features of the IBM Ultrium Tape Drives and the IBM 3580 Ultrium Tape Drive
Ultrium 3
Feature Ultrium 1 Ultrium 2 Ultrium 3 Half-Height Ultrium 4
Ultrium
Native storage
100 GB 200 GB 400 GB 400 GB 800 GB
capacity
Storage capacity
at 2:1 200 GB 400 GB 800 GB 800 GB 1.6 TB
compression
Native sustained
15 MB/s 35 MB/s 80 MB/s 60 MB/s 120 MB/s
data transfer rate
Data transfer rate
at 2:1 30 MB/s 70 MB/s 160 MB/s 120 MB/s 240 MB/s
compression
Burst data
transfer rate 80 MB/s 160 MB/s 160 MB/s 160 MB/s 160 MB/s
(Ultra 160)
Burst data
transfer rate 100 MB/s 100 MB/s 100 MB/s N/A 100 MB/s
(1GFC)
Burst data
transfer rate N/A 200 MB/S 200 MB/S N/A 200 MB/S
(2GFC)
Burst data
transfer rate N/A N/A 400 MB/s N/A 400 MB/s
(4GFC)
Burst data
transfer rate (3G N/A N/A N/A 300 MB/s 300 MB/s
SAS)
Burst data
transfer rate N/A N/A N/A N/A 320 MB/s
(Ultra 320)
Type of interface U160
U2 U160 U160 U160 LC-D
SC-D LC-D LC-D SAS SAS
U320
Note: All sustained data rates are dependent on the capabilities of the interconnect (for example, an UltraSCSI bus
is limited to less than 40MB/sec).
U2: Ultra2 LVD/SE SCSI
U160: Ultra 160 LVD SCSI
U320: Ultra 320 LVD SCSI
SC-D: SC-Duplex Fibre Channel, with the use of SCSI protocol
LC-D: LC-Duplex Fibre Channel, with the use of SCSI protocol
SAS: Serial-Attached SCSI
2 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Supported Servers and Operating Systems
The Ultrium Tape Drives are supported by a wide variety of servers and operating
systems, as well as adapters. These attachments can change throughout the
products’ life cycles. To determine the latest supported attachments, visit the web
at http://www.ibm.com/storage/lto and click on Technical Support or LTO
Support.
Primary Interface Attachment

The Ultrium Tape Drives attach to servers and operating systems shown in Table 2.
An attachment includes (but is not limited to) the servers and operating systems in
the table.
For specific instructions about attachment, see one or more of the following:
v The chapter about installation in the IBM Ultrium Internal Tape Drive Models T200
and T200F Setup, Operator, and Service Guide
v IBM TotalStorage LTO Ultrium 2 Tape Drive Models T400 and T400F Setup, Operator,
and Service Guide
v IBM 3580 Ultrium Tape Drive Setup, Operator, and Service Guide for Models L11,
H11, L13, and H13
v IBM TotalStorage Ultrium Tape Drive 3580 Models L23 and H23 Setup and Operator
Guide
v IBM TotalStorage 3580 Tape Drive Models L33/L3H Setup, Operator, and Service Guide
v IBM TotalStorage Ultrium 3 Tape Drive Setup, Operator, and Service Guide
v IBM TotalStorage Ultrium 4 Tape Drive Setup, Operator, and Service Guide
Table 2. Supported Servers and Operating Systems for Primary Interface Attachment
Supported Servers Supported Operating Systems
IBM AS/400 or ERserver iSeries
® ™
OS/400®
IBM RS/6000®, RS/6000 SP™, or Eserver pSeries® AIX®
IBM Eserver zSeries® 800 or 900 Linux®
Hewlett-Packard HP-UX
Sun Microsystems Solaris
32-bit, Intel-compatible servers Microsoft Windows 2000 or Windows NT®
Red Hat Linux
® ®
64-bit, Intel Itanium servers Red Hat Linux
Supported SAN Components for Fibre Channel Attachment
Visit the web at: http://www.storage.ibm.com/hardsoft/tape/supserver/support.html
Chapter 1. Introduction 3
Supported Device Drivers
IBM maintains the latest levels of device drivers and driver documentation for the
IBM Ultrium Tape Drives on the Internet. You can access this material from your
browser or through the IBM FTP site by performing one of the following
procedures. (Note: If you do not have Internet access and you need information
about device drivers, contact your Marketing Representative.)
v Using a browser, type one of the following:
– http://www.ibm.com/storage
– ftp://ftp.software.ibm.com/storage/devdrvr
– ftp://207.25.253.26/storage/devdrvr
v Using an IBM FTP site, enter the following specifications:
– FTP site: ftp.software.ibm.com
– IP Addr: 207.25.253.26
– Userid: anonymous
– Password: (use your current e-mail address)
– Directory: /storage/devdrvr
IBM provides PostScript- and PDF-formatted versions of its documentation in the

/storage/devdrvr/doc directory:
v IBM_Tape_Driver_IUG.pdf contains the current version of the IBM Tape Device
Drivers Installation and User’s Guide
v IBM_Tape_Driver_PROGREF.pdf contains the current version of the IBM Tape
Device Drivers Programming Reference
Device drivers and utilities for each supported server are beneath
/storage/devdrvr/ in the following directories (the device driver for the iSeries or
AS/400 server is included in the OS/400 operating system):
v AIX
v HPUX
v Linux
v Solaris
v Windows Server 2003
For more information about device drivers, refer to any of the preceding
directories.

Supported Tape Cartridges
The IBM LTO Ultrium Tape Drives support LTO Cartridges as described in Table 3.
Table 3. IBM LTO Ultrium Tape Drive Support of LTO Cartridges
LTO Data Capacity Data Capacity Supported by

Type
Generation Native Compressed Ultrium 4 Ultrium 3 Ultrium 2 Ultrium 1
4 A 800 GB 1600 GB Write and No No No
Read
3 A 400 GB 800 GB Write and Write and No No
Read Read
2 A 200 GB 400 GB Read Only Write and Write and No
Read Read
1 A 100 GB 200 GB No Read Only Write and Write and
Read Read
1 B 50 GB 100 GB No Read Only Write and Write and
Read Read
1 C 30 GB 60 GB No Read Only Write and Write and
Read Read
1 D 10 GB 20 GB No Read Only Write and Write and
Read Read
1. Ultrium Tape Drives reads tapes that have been written by other licensed Ultrium drives.
2. Ultrium Tape Drives write tapes that can be read by other licensed Ultrium drives.
3. Ultrium Tape Drives offer read/write capability for certified LTO Ultrium tape cartridges as specified in this
table.
4. The tape cartridges define the format to which it is written. (for example Ultrium 2 cartridges can only be
written to Ultrium 2 format regardless of which generation drive writes it.
5. Ultrium 3 and Ultrium 4 support WORM cartridges.
The Ultrium 4 Tape Drive (Generation 4) uses the IBM TotalStorage 800 GB Data
Cartridge, and is compatible with the cartridges of its predecessors (called
Generation 2 and Generation 3). The Ultrium 4 Tape Drive performs the following
functions:
v Reads and writes Generation 4 cartridges to Generation 4 format
v Reads Generation 2 cartridges in Generation 2 format
v Does not write Generation 4 cartridges to other generations’ format
v Does not write or read Generation 1 cartridges in any format
The Ultrium 3 Tape Drive (Generation 3) uses the IBM TotalStorage 400 GB Data
Cartridge, and is compatible with the cartridges of its predecessors (called
Generation 1 and Generation 2). The Ultrium 3 Tape Drive performs the following
functions:
v Reads Generation 1 cartridges in Generation 1 format
Chapter 1. Introduction 5
The Ultrium 3 Tape Drive reads tapes that have been written by other licensed
Ultrium 3 drives. It also writes tapes that can be read by other licensed Ultrium 3
drives.
Ultrium 3 Tape Drives offer read/write capability for certified LTO Ultrium tape
cartridges that have capacities of 400 and 200 GB. They also offer read capability
for certified LTO Ultrium tape cartridges that have a capacity of 100 GB.
The Ultrium 2 Tape Drive (Generation 2) uses the IBM TotalStorage LTO Ultrium
200 GB Data Cartridge and is compatible with the cartridges of its predecessor, the
IBM Ultrium Internal Tape Drive (called Generation 1). The Ultrium 2 Tape Drive
performs the following functions:
v Does not write Generation 2 cartridges to Generation 1 format
v Does not write Generation 1 cartridges to Generation 2 format
The Ultrium 2 Tape Drive reads tapes that have been written by other licensed
Ultrium 2 drives. It also writes to tapes that can be read by other licensed Ultrium
2 drives.
cartridges that have capacities of 200 and 100 GB.
cartridges that have a capacity of 100 GB.

Chapter 2. Summary of Drive Generation Differences
This chapter provides a summary of the differences in host attachment protocol
between the Ultrium Internal Tape Drive (Generation 1), the TotalStorage LTO
Ultrium 2 Tape Drive (Generation 2), and the TotalStorage Ultrium 3 Tape Drive
(Generation 3), the TotalStorage Ultrium 3 Half-High Tape Drive (Generation
3HH), and the TotalStorage Ultrium 4 Tape Drive (Generation 4).
The features of the Generation 2 drive that differ from the Generation 1 drive
include:
v 64-MB read-and-write cache
v Speed matching
v Channel calibration
v SET CAPACITY SCSI command
v Ultra160 SCSI interface
v Drive external SCSI termination required
v Fibre Channel 2-Gb/s interface
v Fibre Channel support for direct connection to an F port (for example, a McData
switch)
v Parallel SCSI Ultra 320 interface
The features of the Ultrium 3 Tape Drive that differ from those of the Ultrium 2
Tape Drive include the following:
v 128 MB read-and-write cache
v New media, new media shell color (dark bluish gray, Pantone color number
7546C)
v Expanded request sense length to 96 bytes
v WORM capability
v Fibre Channel 2Gbit/sec interface or 4Gbit/sec interface
v Half-Height drive option with:
1. Serial Attached SCSI (SAS) Interface, or
2. Ultra160 SCSI Interface
The features of the Ultrium 4 Tape Drive that differ from those of the Ultrium 3
HH, Ultrium 3, and Ultrium 2 Tape Drive include the following:
v Half-Height drive option with:
1. Serial Attached SCSI (SAS) Interface, or
2. Ultra160 SCSI Interface
v 128 MB read-and-write cache
v Encryption of data on Ultrium 4 cartridges
1. T10 key management method
2. Transparent management method
a. when using IBM device driver, or
b. when in an IBM library
v New log pages
1. Device Statistics log page (14h)

2. Tape Diagnostic data log page (16h)
3. Performance Characteristics log page (37h)
v Fibre Channel 4Gbit/sec interface
v Serial Attached SCSI (SAS) interface
v Ultra320 SCSI Interface (OEM model only)
v WORM capability
Differences in Command Timeout Values

Due to differences between each of the of the Ultrium drives, the maximum
amount of time it takes for various SCSI commands to execute and return status
has changed. Table 4 provides a list of all recommended host command time-outs
from commands defined by the referenced SCSI-3 standard or by this product as
vendor-unique for sequential access devices. The table lists the following
information for each command: the operation code, recommended timeout, and
notes.
It is strongly recommended that device drivers or host software implement device

reservations using the Reserve or Persistent Reserve commands. Due to the
sequential nature of tape devices, many host commands are serialized, and
command time-outs consequently have an additive effect. Using reservations will
prevent this from causing application disruptions in a multi-initiator or SAN
environment. Similar additive timeout effects can occur if the host is using
command queueing (that is, simple queuing).
Note: The time-outs in the following table are based on the time from the start of
execution of the command, to its reported completion. Since applications are
generally concerned with the time from the command being issued, to its
reported completion, it should be noted that this overall time may be
affected by currently executing operations. Some of these conditions include:
v A prior command was issued with the Immediate bit set in the CDB
v Multiple concurrent commands with Simple queueing are executed
v Multi-initiator configurations without reservations
v Non-host operations, such as manual unloads, power-on self tests, and so
on
v Commands issued shortly after certain aborted commands
v Commands that force flushes when unwritten write data is in the buffer
Table 4. Command Timeout Values (Ultrium 1, 2, and 3 Full-Height) - Sorted by OpCode
Timeout
for Timeout for Ultrium 2 Timeout for Ultrium 3 Full-Height
Ultrium Tape Drive (in minutes) Tape Drive (in minutes)
Tape
Drive
Op (in Gen 1 Gen 2 Gen 1 Gen 2 Gen 3
Code Command minutes) Cartridge Cartridge Cartridge Cartridge Cartridge
00h Test Unit Ready 1 1 1 1 1 1
01h Rewind 8 9 8 8 8 9
03h Request Sense 1 1 1 1 1 1
05h Read Block Limits 1 1 1 1 1 1
08h Read 18 18 18 16 16 17

Table 4. Command Timeout Values (Ultrium 1, 2, and 3 Full-Height) - Sorted by OpCode (continued)
Timeout
Tape
Drive
0Ah Write 18 18 18 N/A 16 18
0Bh Set Capacity N/A 13 13 N/A 11 12
10h Write FileMark 15 15 15 N/A 15 17
11h Space(6) (normal) 16 15 14 14 14 16
Space(6) (slow) 173 138 151 127 165 140
12h Inquiry 1 1 1 1 1 1
13h Verify 18 18 18
15h/55h Mode Select
1 1 1 1 1 1
(6/10)
16h/56h Reserve Unit
1 1 1 1 1 1
(6/10)
17h/57h Release Unit
1 1 1 1 1 1
(6/10)
19h Erase 204 138 151 N/A 160 134
1Ah/5Ah Mode Sense
1 1 1 1 1 1
(6/10)
1Bh Load (Cartridge
11 12 12 8 8 8
Insert -> BOM)
Load (Lp4 ->
8 9 8 8 8 9
BOM)
Unload (BOM ->
10 10 10 10 10 11
Cartridge Eject)
Unload (LP4 ->
11 12 11 11 11 12
Cartridge Eject)
1Ch Receive
1 1 1 1 1 1
Diagnostic Results
1Dh Send Diagnostic 29 35 35 13 39 34
1Eh Prevent/Allow
1 1 1 1 1 1
Medium Removal
2Bh/92h Locate(10/16)
16 15 14 14 14 16
(normal)
Locate(10/16)
173 138 151 127 165 140
(slow)
34h Read Position 1 1 1 1 1 1
3Bh Write Buffer 8 8 8 8 8 8
3Ch Read Buffer 8 8 8 7 7 8
44h Report Density
1 1 1 1 1 1
Support
4Ch Log Select 1 1 1 1 1 1
4Dh Log Sense 1 1 1 1 1 1
Chapter 2. Drive Generation Differences 9

Table 4. Command Timeout Values (Ultrium 1, 2, and 3 Full-Height) - Sorted by OpCode (continued)
Timeout
Tape
Drive
5Eh Persistent Reserve
1 1 1 1 1 1
In
5Fh Persistent Reserve
1 1 1 1 1 1
Out
8Ch Read Attribute 1 1 1 1 1 1
8Dh Write Attribute 1 1 1 1 1 1
91h Space(16)
N/A N/A N/A 14 14 16
(normal)
Space(16) (slow) N/A N/A N/A 127 165 140
A0h Report LUNs 1 1 1 1 1 1
A2h Security Protocol
N/A N/A N/A N/A N/A 1
In (SPIN)
Report Supported
A3h:0Ch N/A N/A N/A 1 1 1
Op Codes
Report Supported
A3h:0Dh Task Management N/A N/A N/A 1 1 1
Functions
A3h:0Fh Report Timestamp N/A N/A N/A 1 1 1
A4h:0Fh Set Timestamp N/A N/A N/A 1 1 1
Security Protocol
B5h N/A N/A N/A N/A N/A N/A
Out (SPOUT)
Table 5. Command Timeout Values (Ultrium 3 Half-Height and Ultrium 4) - Sorted by OpCode
Timeout for Ultrium 3 Half-Height Tape Timeout for Ultrium 4 Tape Drive (in
drive (in minutes) minutes)
Op Code Command
Gen 1 Gen 2 Gen 3 Gen 2 Gen 3 Gen 4
Cartridge Cartridge Cartridge Cartridge Cartridge Cartridge
Test Unit
00h 1 1 1 1 1 1
Ready
01h Rewind 11 11 13 8 9 9
Request
03h 1 1 1 1 1 1
Sense
Read Block
05h 1 1 1 1 1 1
Limits
08h Read 21 21 23 16 17 22
0Ah Write N/A 21 24 N/A 18 23
Write
10h N/A 21 23 N/A 17 22
FileMark

Table 5. Command Timeout Values (Ultrium 3 Half-Height and Ultrium 4) - Sorted by OpCode (continued)
Op Code Command
Space(6)
20 20 22 14 16 21
(normal)
11h
Space(6)
199 264 201 165 140 183
(slow)
12h Inquiry 1 1 1 1 1 1
13h Verify 21 21 23 16 17 22
Mode Select
15h/55h 1 1 1 1 1 1
(6/10)
Reserve Unit
16h/56h 1 1 1 1 1 1
(6/10)
Release Unit
17h/57h 1 1 1 1 1 1
(6/10)
19h Erase N/A 191 255 N/A 134 180
Mode Sense
1Ah/5Ah 1 1 1 1 1 1
(6/10)
Load
(Cartridge
9 9 9 8 8 8
Insert ->
BOM)
Load (LP4 ->
11 11 13 8 9 9
BOM)
1Bh Unload
(BOM ->
12 12 14 10 11 11
Cartridge
Eject)
Unload (LP4
-> Cartridge 14 14 16 11 12 13
Eject)
Receive
1Ch Diagnostic 1 1 1 1 1 1
Results
Send
1Dh 13 39 40 13 34 35
Diagnostic
Prevent/
Allow
1Eh 1 1 1 1 1 1
Medium
Removal
Locate(10/
20 20 22 14 16 21
16) (normal)
2Bh/92h
Locate(10/
199 264 201 165 140 183
16) (slow)
Read
34h 1 1 1 1 1 1
Position

Table 5. Command Timeout Values (Ultrium 3 Half-Height and Ultrium 4) - Sorted by OpCode (continued)
Op Code Command
Report
44h Density 1 1 1 1 1 1
Support
Persistent
5Eh 1 1 1 1 1 1
Reserve In
Persistent
5Fh 1 1 1 1 1 1
Reserve Out
Space(16)
20 20 22 14 16 21
(normal)
91h
Space(16)
199 264 201 165 140 183
(slow)
Read
BCh 1 1 1 1 1 1
Attribute
Write
BDh 1 1 1 1 1 1
Attribute
Security
A2h N/A N/A N/A N/A N/A 1
Protocol In
Report
A3h:0Ch Supported 1 1 1 1 1 1
Op Codes
Report
Supported
A3h:0Dh Task 1 1 1 1 1 1
Management
Functions
Report
A3h:0Fh 1 1 1 1 1 1
Timestamp
Set
A4h:0Fh 1 1 1 1 1 1
Timestamp
Security
B5h N/A N/A N/A N/A N/A 1
Protocol Out
Table 6. Command Timeout Values (Ultrium 3 Half-Height and Ultrium 4) - Alphabetical

Op Code Command
19h Erase N/A 191 255 N/A 134 180
12h Inquiry 1 1 1 1 1 1

Table 6. Command Timeout Values (Ultrium 3 Half-Height and Ultrium 4) - Alphabetical (continued)
Op Code Command
Load
(Cartridge
9 9 9 8 8 8
Insert ->
1Bh BOM)
Load (LP4 ->
11 11 13 8 9 9
BOM)
Locate(10/
20 20 22 14 16 21
16) (normal)
2Bh/92h
Locate(10/
199 264 201 165 140 183
16) (slow)
Mode Select
15h/55h 1 1 1 1 1 1
(6/10)
Mode Sense
1Ah/5Ah 1 1 1 1 1 1
(6/10)
Persistent
5Eh 1 1 1 1 1 1
Reserve In
Persistent
5Fh 1 1 1 1 1 1
Reserve Out
Prevent/
Allow
1Eh 1 1 1 1 1 1
Medium
Removal
Report 1 1 1 1 1 1
A3h:0Ch Supported
Op Codes
Report 1 1 1 1 1 1
Supported
A3h:0Dh Task
Management
Functions
Report 1 1 1 1 1 1
A3h:0Fh
Timestamp
Request 1 1 1 1 1 1
03h
Sense
Reserve Unit 1 1 1 1 1 1
16h/56h
(6/10)
08h Read 21 21 23 16 17 22
Read
BCh 1 1 1 1 1 1
Attribute
Read Block
05h 1 1 1 1 1 1
Limits

Table 6. Command Timeout Values (Ultrium 3 Half-Height and Ultrium 4) - Alphabetical (continued)
Op Code Command
Read
34h 1 1 1 1 1 1
Position
Receive
1Ch Diagnostic 1 1 1 1 1 1
Results
Release Unit
17h/57h 1 1 1 1 1 1
(6/10)
Report
44h Density 1 1 1 1 1 1
Support
01h Rewind 11 11 13 8 9 9
Security N/A N/A N/A N/A N/A 1
A2h Protocol In
(SPIN)
Security N/A N/A N/A N/A N/A 1
B5h
Protocol Out
Send
1Dh 13 39 40 13 34 35
Diagnostic
Set
A4h:0Fh 1 1 1 1 1 1
Timestamp
Space(6)
20 20 22 14 16 21
(normal)
11h
Space(6)
199 264 201 165 140 183
(slow)
Space(16)
20 20 22 14 16 21
(normal)
91h
Space(16)
199 264 201 165 140 183
(slow)
Test Unit
00h 1 1 1 1 1 1
Ready
Unload
(BOM ->
12 12 14 10 11 11
Cartridge
1Bh Eject)
Unload (LP4
-> Cartridge 14 14 16 11 12 13
Eject)
13h Verify 21 21 23 16 17 22
0Ah Write N/A 21 24 N/A 18 23
Write
BDh 1 1 1 1 1 1
Attribute
Write
10h N/A 21 23 N/A 17 22
FileMark

Command and Parameter Differences Between Generations
Table 7 shows commands and parameters added since Ultrium 1 and in which
generation(s) it is applicable.
Table 7. Command and Parameter differences between generations
Command or Generation
Description
Parameter 2 3 4
Set Capacity The SET CAPACITY command is supported. The minimum capacity Y Y Y
Command allowed varies from one generation to another. For more information, see
“SET CAPACITY (not Ultrium 1)” on page 248.
Echo Buffer Support Echo Buffer mode of the READ BUFFER and WRITE BUFFER Y Y Y
commands (see “READ BUFFER” on page 173 and “WRITE BUFFER” on
page 258, respectively).
Volume Identifier Volume Identifier Part of the Device Attribute of Medium Auxiliary - - Y
Memory. This is the volume identifier (VolSer or BarCode) value set by
the library
Security Protocol In This command is related to managing encryption. See “SECURITY - - Y
(SPIN) PROTOCOL IN (SPIN) A2h” on page 201.
Security Protocol Out This command is related to managing encryption. See “SECURITY - - Y
(SPOUT) PROTOCOL OUT (SPOUT) B5h” on page 215.
Key:
- Not Supported
Data Changes Between Generations

The sections that follow describe the data changes between Generations of the
Ultrium tape drive.
Standard Inquiry Data

The length of Standard Inquiry data on Ultrium 1 drives is less than on later
generations. The Standard Inquiry data reported by Ultrium drives is shown in
Table 11 on page 22. The Additional Length field specifies how many bytes are
returned. Currently, Ultrium 1 devices set this value to 33 (21h) and later
generation devices set this value to 53 (35h). This value is subject to change. It is
strongly recommended that the user parse the data returned by using the
Additional Length field instead of the published values.
Product Identification information returned is shown in Table 13 on page 26.
For more information, see “Standard Inquiry Data” on page 22.
REPORT DENSITY SUPPORT Command

The REPORT DENSITY SUPPORT command has added density values for all
generations through Ultrium 4. Additionally, the units in which the Capacity value
are returned has changed to match the SCSI standards. See Table 178 on page 187.
For more information, see “REPORT DENSITY SUPPORT” on page 185.

Mode Pages
On Mode Sense commands the value returned in the Medium Type field is shown
in Table 95 on page 110. On Mode Select commands any value is allowed and
ignored.
Fibre Channel Port Control Page (19h) Page Length changed from 0Eh in Ultrium 1
drives to 06h in later generation 2 drives. This matches the definition in the SCSI
standards (FCP-2 and later). For more information, see “Mode Page 19h: Protocol
Specific Port Control Page” on page 124.
READ POSITION Command

The READ POSITION command for all generation drives except Ultrium 1 behaves
as described in the SCSI standards (SPC-2). For more information, see “READ
POSITION” on page 180.
Request Sense Changes

For Request Sense data, the Additional Sense Length is set to n-7, and is at least
10. When the sense data is associated with an Illegal Length read, the Additional
Sense Length may be 10. In Ultrium 1 and Ultrium 2 drives, n can be as large as
35. In Ultrium 3 and later generation drives, n can be as large as 95. While this
length is not anticipated to change, it is recommended that the Additional Sense
Length be used to parse the data. For more information, see “REQUEST SENSE”
on page 194.
Behavior Changes Between Generations
Cartridge Eject for Errors

Ultrium 1 drives auto-eject data cartridges when errors occur during loads. Later
generation drives do not.
Queueing Issues
For all generations of drive later than Ultrium 1, when a cartridge is inserted into
the drive through means other than SCSI commands to LUN 0, it is assumed that
the host will poll the drive with TEST UNIT READY commands to determine its
readiness before issuing in-order commands (for example, commands other than
INQUIRY, TEST UNIT READY, REQUEST SENSE, or REPORT LUNS). If this is not
the case, these commands may time out in ERP (Error Recovery Procedure)
situations.
Microcode Detection of Errors

The drive microcode is designed to check for logic errors, to handle
hardware-detected errors, and to detect and report microcode-related errors.
Fencing Behavior
For a description of the Fencing Behavior and Persistent Error handling, see
“Persistent Errors” on page 262.

Chapter 3. Command Support
In the sections that follow, each SCSI command includes a table that describes the
fields in the Command Descriptor Block (CDB). The table is similar to those
published by the InterNational Committee for Information Technology Standards
(INCITS). It includes bit numbering conventions that conform to ANSI standards.
The conventions are as follows:
v Bit 0 is the least significant bit (LSB) and occupies the right bit position in the
table
v Bits 1-6 continue from right to left in ascending order
v Bit 7 is the most significant bit (MSB) and occupies the left bit position in the
table
The LUN field in the CDB has been obsoleted in SCSI-3 and is ignored for every
command.
Notes:
1. For this chapter, a megabyte (MB) is equal to 1 048 576 bytes.
2. Binary numbers are represented by numbers followed by b. Hexadecimal
numbers are represented by 0-9 and A-F followed by h. Numbers with no suffix
can be assumed to be decimal.
Table 8. Supported Common SCSI Commands
Applicable Conditions2
Command Name Operation Code SCSI Spec1 Page
RVC UAT NRD WRP MFC DCC
ERASE 19h SSC 20 y y y y y y
INQUIRY 12h SPC-2 21 - - - - - -
LOAD/UNLOAD 1Bh SSC 45 y y - - y y
LOCATE(10) 2Bh SSC 47 y y y - y y
LOCATE(16) 92h SSC-3 47 y y y - y y
LOG SELECT 4Ch SPC-2 49 y y - - - y
LOG SENSE 4Dh SPC-2 50 - - - - - -
MODE SELECT (6) 15h SPC-2 106 y y - - - y
MODE SELECT (10) 55h SPC-2 106 y y - - - y
MODE SENSE (6) 1Ah SPC-2 107 - y - - - -
MODE SENSE (10) 5Ah SPC-2 107 - y - - - -
PERSISTENT RESERVE IN 5Eh SPC-2 144 y y - - - -
3
PERSISTENT RESERVE 5Fh SPC-2 149 y y - - - -
OUT
PREVENT ALLOW 1Eh SPC-2 153 - y - - - -
MEDIUM REMOVAL
READ 08h SSC 154 y y y - y y
READ ATTRIBUTE 8Ch SPC-3 156 y y - - - y
READ BLOCK LIMITS 05h SSC 172 - y - - - -
READ BUFFER 3Ch SPC-2 173 y - - - - -
READ POSITION 34h SSC 180 y y - - - -
RECEIVE DIAGNOSTIC 1Ch SPC-2 183 y y - - - -
RESULTS
RELEASE UNIT (6) 17h SPC-2 184 - y - - - -

Table 8. Supported Common SCSI Commands (continued)
RELEASE UNIT (10) 57h SPC-2 184 - y - - - -
REPORT DENSITY 44h SSC 185 - y - - - -
SUPPORT
REPORT LUNs A0h SSC 190 - - - - - -
A3h/0Dh SPC-4 “REPORT - - - - - -
SUPPORTED
REPORT SUPPORTED TASK
TASK MANAGEMENT MANAGEMENT
FUNCTIONS FUNCTIONS
(Not LTO1)” on
page 192
REQUEST SENSE 03h SPC-2 194 - - - - - -
RESERVE UNIT (6) 16h SPC-2 199 y y - - - -
RESERVE UNIT (10) 56h SPC-2 199 y y - - - -
REWIND 01h SSC 200 y y y - y y
SECURITY PROTOCOL IN A2h SPC-4 201 y y - - - -
(SPIN)
SECURITY PROTOCOL B5h SPC-4 215 y y - - y -
OUT (SPOUT)
SEND DIAGNOSTIC 1Dh SPC-2 223 y y - - y y
0Bh SSC-2 “SET CAPACITY y y y y y y
SET CAPACITY (not Ultrium 1)”
on page 248
SPACE 11h SSC 252 y y y - y y
91h SSC-3 Table 249 on y y y - y y
SPACE (16)
page 252
TEST UNIT READY 00h SSC 254 y y y - - y
VERIFY 13h SSC 255 y y y - y y
WRITE 0Ah SSC 256 y y y y y y
WRITE ATTRIBUTE 8Dh SPC-3 257 y y y y - y
WRITE BUFFER 3Bh SPC-2 256 y y - - - -
WRITE FILE MARKS 10h SSC 260 y y y y y y

Table 8. Supported Common SCSI Commands (continued)
Note 1: SCSI specifications are as follows:
SSC-3 = SCSI Stream Commands - 3
SSC-2 = SCSI Stream Commands - 2
SSC = SCSI-3 Stream Commands
SPC-2 = SCSI Primary Commands-2
Note 2: Applicable Conditions are as follows:

y = condition can apply to the command
- = condition cannot apply to the command
RVC = reservation conflict
UAT = unit attention
NRD = not ready
WRP = write protect
MFC = medium format corrupted. These commands are also considered Medium Access Commands
DCC = deferred check condition
Note 3: Reported as appropriate for the type of Service Action and Reservation Type requested and the current reservation state of
the drive.
Note 4: If the drive sets the CRH bit to one in the parameter data returned by the REPORT CAPABILITIES service action of the
PERSISTENT RESERVE IN command, then in the presence of a Persistent Reservation, a RELEASE(6) or RELEASE(10) command
will complete with GOOD status, but the persistent reservation will not be released, if the command is received from:
1. An I_T nexus that is a persistent reservation holder; or
2. An I_T nexus that is registered if a registrants only type persistent reservation is present.
In all other cases, the command will be processed as defined in SPC-2.
Note 5: A RESERVE(6) or RESERVE(10) command will complete with GOOD status, but no reservation will be established and the
persistent reservation will not be changed, if the command is received from:
1. An I_T nexus that is a persistent reservation holder; or
2. An I_T nexus that is registered if a registrants only type persistent reservation is present.
In all other cases, the command will be processed as defined in SPC-2.
Chapter 3. Command Support 19

ERASE
Table 9. ERASE Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (19h)
1 Obsolete Reserved Immed Long
2 Reserved
3 Reserved
4 Reserved
5 Control
EOD is written at the current position, which marks it as end of data.
If the Long field is set to 0, no further writing occurs. If the Long field is set to 1,
the Data Set Separator (DSS) pattern is written from EOD to the end of the
medium to overwrite any data that is currently on the tape.
If the Immediate (Immed) field is set to 1, then the drive validates the command
and waits for any previous command from any server to complete, including any
immediate commands that are currently being processed. It also waits for any
buffered data to be flushed to tape. It then reports a deferred error for any
preceding command or buffered data, if appropriate. If there is no deferred error,
the drive reports Good status and initiates the command. If the Immediate
(Immed) field is set to 0, status is not returned until after the command has
completed.
After the command is successfully completed, the drive is positioned immediately

before End Of Data (not End Of Tape).

INQUIRY
The INQUIRY command instructs the drive to return data about itself to the
initiator.
Table 10. INQUIRY Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved EVPD
2 Page Code
3 Reserved
4 Allocation Length
5 Control
If the Enable Vital Product Data (EVPD) field is set to 0 and the Page Code is 0,
Standard Inquiry Data is returned.
If the Enable Vital Product Data (EVPD) field is set and the Page Code is 0, the
Supported Vital Product Data Pages page is returned. This page lists the EVPD
pages that are supported by the drive in this configuration.
If the Enable Vital Product Data (EVPD) field is set, if the Page Code is not 0, and
if there is an Inquiry data page that corresponds to that page code, then that page
is returned. See Table 14 on page 27 for supported data pages. The contents of
page D0h are not specified in this document.
If the preceding conditions do not apply, Check Condition status is returned. The
Sense Key is set to Illegal Request (5) and the ASC/ASCQ is set to Invalid Field in
CDB (2400).

Standard Inquiry Data
In drives that are not configured with the eServer™ attachment feature, Standard
Inquiry Data for a valid LUN is described in Table 11.
In drives configured with the eServer attachment feature, the Standard Inquiry
Data for a valid LUN is defined in Table 12 on page 24.
WARNING
Drives that have eServer Attachment Enabled have several behaviors that are
different from the drives without eServer Attachment Enabled. Please see the
following sections and look for eServer attachment differences: “Mode Page
2Fh: Behavior Configuration Mode Page” on page 134; “REWIND” on page
200; “WRITE FILE MARKS” on page 260.
Table 11. Standard Inquiry Data Valid LUN

Bit
Byte 7 6 5 4 3 2 1 0
0 Peripheral Qualifier Peripheral Device Type
1 RMB Reserved
2 ISO version (00b) ECMA version (000b) Version (3h)
3 AERC(0) Obsolete NACA(0) HiSup(0) Response Data Format (2h)
4 Additional Length
5 Reserved
6 BQue(0) EncSrv(0) VS(0) MultiP MChngr(0) Obsolete Adr32(0) Adr16
7 RelAdr(0) Obsolete WBs16 Sync Linked(0) TransDis(0) CmdQ VS(0)
8
Vendor Identification
:
IBM
15
16
: Product Identification
31
32
: Product Revision Level: YMDV
35
36 Reserved AutDis
37 Performance Limit
38
: Reserved
40
41 OEM Specific

Table 11. Standard Inquiry Data Valid LUN (continued)
Bit
Byte 7 6 5 4 3 2 1 0
42
: Reserved
55
56 Reserved Clocking QAS (0b) IUS
57 Reserved

Table 12. Standard Inquiry Data Valid LUN (Drive configured with eServer attachment feature)
Bit
Byte 7 6 5 4 3 2 1 0
0 Peripheral Qualifier Peripheral Device Type
1 RMB Reserved
2 ISO version (00b) ECMA version (000b) Version (3h)
3 AERC(0) Obsolete NACA(0) HiSup(0) Response Data Format (2h)
4 Additional Length
5 Reserved
6 BQue(0) EncSrv(0) VS(0) MultiP MChngr(0) Obsolete Adr32(0) Adr16
7 ReAdr(0) Obsolete WBs16 Sync Linked(0) TransDis(0) CmdQ VS(0)
8 Vendor Identification (in left-aligned ASCII)
“IBM”
15
16 Product Identification (in left-aligned ASCII)
31
32 Product Revision Level (in left-aligned ASCII): YMDV
35
36 Unit Serial Number (8 ASCII characters)
[″20ssssss″ = last 6 characters of the 11S identifier]
43
44
Load ID (pSCSI: A1700D63h SAS: A1700D64h)
47
IBM Drive Part Number (8 ASCII characters) [The first 7 bytes of this field
48 match the 7-character drive P/N printed on the rear drive label]
i/p Series pSCSI “23R7031ø”
55 i/p Series SAS “23R7037ø”
x Series SAS “23R7035ø”
Clocking
56 Reserved pSCSI: (11b) QAS (0b) IUS (0b)
SAS: (00b)
57
Reserved
95
96
Reserved
97
98
Plant ID (4 ASCII characters) (“N1øø”)
101
102
Reserved
112
113 Reserved LIBR(0)
114 Drive FRU Part Number (12 ASCII characters)

i/p Series pSCSI “23R7032øøøøø”
i/p Series SAS “23R7038øøøøø”
125 x Series SAS “23R7036øøøøø”
126 IBM EC Level (10 ASCII characters)) [The first 6 bytes of this field match the EC
level printed on the rear drive label]
135 i/p/x Series “H82204øøøø”

For Valid LUN:
v Peripheral Qualifier is set to 000b.
v Peripheral Device Type is set to 01h.
v RMB is set to 1b to indicate that the medium is removable.
For Invalid LUN:
v Peripheral Qualifier is set to 011b.
v Peripheral Device Type is set to 1Fh.
v RMB is set to 0b to indicate that the medium is not removable.
For Ultra 160 SCSI device:
v Adr16 field is set to 1, which indicates that the drive supports 16 SCSI IDs.
v WBs16 field is set to 1, which indicates that the drive supports a 16-bit wide
data path on a single cable.
v Sync field is set to 1, which indicates that the drive supports synchronous data
transfers.
v CmdQ field is set to 0, which indicates that the drive does not command
queueing.
v Clocking field is supported on Ultrium 2 and later drives only and is set to 11b
because the drive supports both ST and DT modes. On Ultrium 1 devices, this
field is set to 00b.
v QAS field is set to 0, which indicates that the drive does not support quick
arbitration and selection.
v IUS field is set to 1, which indicates that the drive supports information unit
transfers.
For Ultra 320 SCSI device:

v Adr16 field is set to 1, which indicates that the drive supports 16 SCSI IDs.
v WBs16 field is set to 1, which indicates that the drive supports a 16-bit wide
data path on a single cable.
v Sync field is set to 1, which indicates that the drive supports synchronous data
transfers.
v CmdQ field is set to 1, which indicates that the drive supports tagged (simple)
command queueing.
v Clocking field is set to 11b, which indicates that the drive supports both ST and
DT modes.
v IUS field is set to 1, which indicates that the drive supports information unit
transfers. For Fibre Channel and SAS devices:
For Fibre Channel and SAS devices:

v Adr16 field is set to 0.
v WBs16 field is set to 0.
v Sync field is set to 0.
v CmdQ field is set to 1, which indicates that the drive supports tagged (simple)
command queueing.
v Clocking field is set to 0.

v IUS field is set to 0, which indicates that the drive does not support information
unit transfers.
For all devices:

v The Additional Length field specifies how many bytes follow. Currently Ultrium
1 devices set this value to 33 (21h), and Ultrium 2 and later devices set this
value to 53 (35h). This value is subject to change. It is strongly recommended
that the user parse the data returned by using the Additional Length field
instead of the published values.
v Vendor Identification returns IBM in ASCII with trailing blanks.
v Product Vendor Identification returns ULTxxxx-TDy in ASCII with trailing blanks.
If the drive is an IBM drive, xxxx equals 3580; if it is an OEM drive, xxxx equals
RIUM. The character y indicates the generation of the drive. See the table below
for the value returned.
Table 13. Product Identification Table
Generation IBM Drive OEM Drive eServer Attachment Enabled
1 ULT3580-TD1 ULTRIUM-TD1 N/A
3 Half-height ULT3580-HH3 ULTRIUM-HH3 HH LTO Gen 3
4 Half-height ULT3580-HH4 ULTRIUM-HH4 HH LTO Gen 4
Note: IBM Tape Device Drivers will only configure Product IDs that are found in the columns entitled "IBM Drive"
and "OEM Drive".
Note: IBM SCDD will only configure Product IDs that are found in the column entitled "eServer Attachment
Enabled".
v Product Revision Level has four parts:

– Y is the last character of the year (for example, 2 indicates the year 2002)
– M is the month, in the alphanumeric set 1 through 9, A, B, or C
– D is the day, in the alphanumeric set 1 through 9, A through V
– V is the version, in the alphanumeric set 0 through 9, a through z, and A
through Z, with 0 being the earliest and Z the latest (to avoid interpretation
errors, the characters i, l, and 0 are not used)
v If the Performance Limit field is set to 0, the drive is a full performing drive. If
the Performing Limit field is set to a non-zero value, the Limit of the drive is
defined by the following equation:
Limit in MB/s = (Max Performance in MB/s) * ((Performance Limit) / 256)
v Automation Disabled (AutDis) field set indicates that the drive is not capable of
full automation function. When this field is 0, it indicates that the drive is
capable of full automation function.
v The OEM Specific field is intentionally not specified. See the documentation
from the OEM vendor to determine the field’s definition and application.

Supported Vital Product Data Page
Table 14. Supported Vital Product Data Inquiry Page
Bit
Byte 7 6 5 4 3 2 1 0
0 Peripheral Qualifier (000b) Peripheral Device Type (01h)
1 Page Code (00h)
2 Reserved
3 Page Length (n-3)
supported VPD page list
4 Supported Vital Product Data Page Code (00h)
5 Firmware Designation Page Code (03h)
6 Unit Serial Number Page Code (80h)
7 Device Identification Page Code (83h)
8 Mode Page Policy Page Code (87h)
9 SCSI Ports Page Code (88h)
(Not returned on U160/U320 attached drives)
10 Sequential-Access Device Capabilities Page Code (B0h)
11 Drive Component Revision Levels Page Code (C0h)
12 Drive Serial Numbers Page (C1h)
13 Device Unique Configuration Page (C7h)
14 Mode Parameter Default Settings Page (C8h)
15 Vendor-Unique ITD Page (D0h/E0h)
16 Vendor-UniqueITD Page (D1h/E1h)
The Supported Vital Product Data Page contains pages that the device will return.
Only those drives that have the AS/400 (iSeries) attachment enabled will have
valid data in pages D0h/D1h/E0h/E1h. The contents of pages D0h/D1h/E0h/E1h
are not specified in this document. OEM drives may add pages not specified in
this document.
The Page Length field indicates how many bytes follow.

Firmware Designation Page (03h)
The Firmware Designation Page is used to identify which code image can be
downloaded to which drive. See Chapter 10, “Firmware Download,” on page 293
for Firmware Download procedures.
Table 15. Firmware Designation Page
Bit Byte 7 6 5 4 3 2 1 0
1 Page Code (03h)
2 Reserved
3 Page Length (21h)
4
Reserved
7
8 Load ID
(Identifies downloadable firmware
11 with the drive level hardware)
12 Firmware Revision Level
15 (Defined in Standard Inquiry Data, bytes 32-35)
16
PTF Number (00000000h)
19
20
Patch Number (00000000h)
23
24 RU Name
31 (8-byte EBCDIC representation of the RU name.)
32
Library Sequence Number
36
The Load ID and RU Name and which drive level hardware they designate are
defined in Table 277 on page 294.

Mode Page Policy Page (87h)
The Mode Page Policy Page (see Table 16) indicates which mode page policy is in
effect for each mode page supported by the logical unit.
Table 16. Mode Page Policy page
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Code (87h)
2
Page Length (n-3)
3
Mode page policy descriptor list
4
: Mode page policy descriptor (first)
7
:
n-3
: Mode page policy descriptor (last)
Each mode page policy descriptor (see Table 17) contains information describing
the mode page policy for one or more mode pages or subpages. The information in
the mode page policy descriptors in this page describe the mode page policy for
every mode page and subpage supported by the logical unit.
Table 17. Mode Page policy descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0 Reserved Policy Page Code
1 Policy Subpage Code
2 MLUS Reserved Mode Page Policy
3 Reserved
The Policy Page Code field and Policy Subpage Code field indicate the mode page
and subpage to which the descriptor applies.
The first mode page policy descriptor in the list contains a Policy Page Code field
set to 3Fh and a Policy Subpage Code field set to FFh, indicating that the
descriptor applies to all mode pages and subpages not described by other mode
page policy descriptors. The Policy Page Code field is set to 3Fh and the policy
Subpage Code field is set to FFh only in the first mode page policy descriptor in
the list.
If the Policy Page Code field contains a value other than 3Fh or a policy Subpage
Code field contains a value other than FFh, then the Policy Page Code field and
the Policy Subpage Code field indicate a single mode page and a subpage to which
the descriptor applies.
A Multiple Logical Units Share (MLUS) bit set to 1 indicates the mode page and
subpage identified by the Policy Page Code field and policy Subpage Code field is
shared by more than one logical unit. An MLUS bit set to zero indicates the logical
unit maintains its own copy of the mode page and subpage identified by the
Policy Page Code field and Policy Subpage Code field. The MLUS values are
described in Table 19
The Mode Page policy field (see Table 18) indicates the mode page policy for the
mode page and subpage identified by the Policy Page Code field and policy
Subpage Code field. The mode page policies are described in Table 18.
Table 18. Mode page policy values
Value Description
00b Shared
01b Per target port
10b Per initiator port
11b Per I_T nexus
Table 19. Mode Page Policy for LUN 0 (SSC Device)

Subpage
Page Name Page Code Code MLUS Mode Page Policy
Read-Write Error Recovery Page 01h 00h 0 Shared (00b)
Disconnect/Reconnect Page 02h 00h 1 Per I_T nexus (11b)
Control Extension Mode Page 0Ah 01h 1 Shared (00b)
Data Compression Mode Page 0Fh 00h 0 Shared (00b)
Sequential Access Device Configuration Page 10h 00h 0 Shared (00b)
Fibre Channel Logical Unit Control Page 18h 00h 0 Per I_T nexus (11b)
Fibre Channel Port Control Page 19h 00h 1 Per target port (01b)
Information Exceptions Mode Page 1Ch 00h 0 Shared (00b)
For LUN 0, this would lead to returning a Mode Page Policy Descriptor of page
3Fh and subpage FFh stating MLUS (0) and Mode Page Policy (000b - Shared).
This would be followed by descriptors for the Disconnect/Reconnect Page, the
Control Extension Mode Page, the Fibre Channel Logical Unit Control Page, and
the Fibre Channel Port Control Page.
Table 20. Mode Page Policy LUN 2 (ADC Device)
Subpage
Mode Page Policy
Page Name Page Code Code MLUS
Control Extension Mode Page 0Ah 01h 1 Shared (00b)
ADC Device Configuration mode page
0Eh 01h 0 Shared (00b)
Node descriptor subpage
DTD Primary Port descriptor subpage
Target Device Serial Number subpage

On LUN 2, this would lead to returning a Mode page Policy Descriptor of page
3Fh and subpage FFh stating MLUS (0) and Mode Page Policy (00b -Shared). This
would be followed by descriptors for the Control Extension Mode Page.

SCSI Ports Page (88h)
The SCSI Ports Inquiry page (see Table 21) is supported on Fibre Channel and
Serial Attached SCSI drives only. It provides a means to retrieve identification
descriptors for all the SCSI ports in the drive.
Table 21. SCSI Ports Inquiry Page
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Code (88h)
2 Reserved
3 Page Length (n-3)
Identification descriptor list
4
First SCSI port identification descriptor
:
(see Table 22 on page 33)
n
:
n
Last SCSI port identification descriptor
:
(see Table 22 on page 33)
m
Each SCSI Port identification descriptor (see Table 22 on page 33) identifies a SCSI
port. The SCSI port identification descriptors may be returned in any order. There
will be one SCSI Port Identification Descriptor for each port in the drive.

Table 22. SCSI Port Identification Descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0
Reserved (0)
1
2
Relative Target Port (see Table 33 on page 41)
3
4
Reserved (0)
5
6
Initiator Port TranscriptID Length (0)
7
8
Reserved (0)
9
10
Target Port Descriptors Length (12)
11
Target port descriptor list
12
: Target port descriptor (see Table 23)
23
Table 23. Target Port Descriptor Format

Bit
Byte 7 6 5 4 3 2 1 0
12 Protocol Identifier Code Set (1)
13 PIV (1) Rsvd Association Type (01b) Identifier Type (3)
14 Reserved
15 Identifier Length (8)
16
: World Wide Port Name (WWPN)
23
The Protocol Identifier value is 0h on Fibre Channel devices and 6h on SAS

devices.

Sequential-Access Device Capabilities Page (B0h)
Table 24. Sequential-Access Device Capabilities Page
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Code (B0h)
2
Page Length (2)
3
4 Reserved WORM
5 Reserved
The Write Once Read Many (WORM) bit is set to 1 on Ultrium 3 and later drives
indicating the drive supports WORM mode operation. The WORM bit is set to 0
on Ultrium 2 drives indicating the drive does not support WORM mode operation.

Drive Serial Numbers (C1h)
Table 25. Inquiry Page C1h: Drive Serial Numbers
Bit Byte 7 6 5 4 3 2 1 0
1 Page Code (C1h)
2 Reserved
3 Page Length (18h)
4
Manufacturing Serial Number, right-justified with leading zeroes, in ASCII
:
Serial Number of device set during manufacturing
15
16
Reported Serial Number, right-justified with leading zeroes, in ASCII
:
Serial Number of device reported over the primary interface (for example Inquiry pages 80h and 83h).
27

Device Unique Configuration Data Page (C7h)
The Device Unique Configuration Data Page is supported on IBM eServer capable
drives only. On drives that are not eServer capable this page will be returned with
all zero data.

Mode Parameter Default Settings Page (C8h)
This page is used on drives that are eServer capable. On drives that are not eServer
capable this page will be returned with all zero data. The Mode Parameter Default
Settings Page is used by sending the Mode Page data starting at byte 14 as the
parameter data of a Mode Select command to the drive to set the default values
required by AIX to use the drive as an internal device.

Unit Serial Number Page (80h)
Table 26. Unit Serial Number Inquiry Page
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Code (80h)
2 Reserved
3 Page Length (0Ah)
4
: Unit Serial Number
13
The Unit Serial Number Page contains a single value that is a 10-byte ASCII string.
The string, with the Vendor Identification and Product Identification fields in the
Standard Inquiry Data, uniquely identifies the drive. Valid Serial Number values
are 0 through 9, A through D, and F. On eServer enabled drives this is the last 10
characters of the 11S identifier

Device Identification Page (83h)
Table 27. Device Identification Inquiry Page
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Code (83h)
2 Reserved
3 Page Length
SCSI drives: (n-3)
Fibre Channel drives: (m-3)
SAS drives (q-3)

4
: Device Identification Descriptor
n
n+1
: Protocol Identification descriptor
Parallel SCSI devices return only the Device Identification Descriptor (see Table 28).
% Fibre Channel and SAS devices return the Device Identification Descriptor (see
Table 28) followed by the Protocol Identification Descriptor (see Table 30 on page
% 40). SAS devices also return the Target Device Name Descriptor (Table 35 on page
% 43).
Table 28. Device Identification Descriptor Format
Bit
Byte 7 6 5 4 3 2 1 0
4 Reserved Code Set (2h)
5 Reserved Identifier Type
6 Reserved
7 Identifier Length (n-7)
8
: Identifier
Table 29 on page 40 describes the Identifier format.

Table 29. Identifier Format
Bit
Byte 7 6 5 4 3 2 1 0
8
: Vendor Identification
15
16
: Product Identification
31
32
: Serial Number
41
The Code Set field is 2, indicating the Identification Descriptor contains only ASCII
data.
This device supports an Identifier Type of 1. In this case, the Device Identification
Descriptor is the Vendor Identification followed by the Product Identification field
from the Standard Inquiry Data and the Serial Number field from the Unit Serial
Number Inquiry Page (see “Unit Serial Number Page (80h)” on page 38).
The Parallel SCSI drive reports only the single Identification Descriptor of
Identifier Type 1.
The Protocol Identification descriptor is returned only on Fibre Channel and SAS
attached devices. The format is given in Table 30.
Table 30. Protocol Identification descriptor
Bit
Byte 7 6 5 4 3 2 1 0
42
: Logical Unit Descriptor
53
54
: Port Identifier Descriptor
61
62
: Port Name (NAA) Descriptor
73

Table 31. Logical Unit (NAA) Descriptor Format
Bit
Byte 7 6 5 4 3 2 1 0
42 Protocol Identifier Code Set (1h)
43 PIV (1) Rsvd Association Type (00b) Identifier Type (3h)
44 Reserved
46
: World Wide Node Name (WWNN)
53

devices.
Table 32. Port Identifier Descriptor Format
Bit
Byte 7 6 5 4 3 2 1 0
54 Protocol Identifier Code Set (1h)
56 Reserved
58
: Obsolete (0)
59
60
: Relative Target Port
61

devices.
Relative Target Port is the Relative Target Port through which the command was
received. Relative Target Port values are defined in Table 33. The port identifier
descriptor is only returned on Ultrium 3 and later drives.
Table 33. Relative Target Port Values
Value Description
0000h Reserved
0001h Relative port 1, historically known as port A or FC port 0
0002h Relative port 2, historically known as port B or FC port 1
0003h Relative port 3, RS422 port
0004h - FFFFh Relative port 4 through 65 535

Table 34. Port Name (NAA) Descriptor Format
Bit
Byte 7 6 5 4 3 2 1 0
62 Protocol Identifier Code Set (1)
64 Reserved
66
: World Wide Port Name (WWPN)
73

devices.
World Wide Port Name (WWPN) is the WWPN through which the command was
received. The port name descriptor is only returned on Ultrium 3 and later drives.

Table 35. Target Device Name Descriptor
Bit
Byte 7 6 5 4 3 2 1 0
74 Protocol Identifier (6h) Code Set(1h)
75 PIV(1) Rsvd Association Type (10b) Identifier Type (3h)
76 Reserved
78
SAS address of drive in NAA IEEE Registered format
:
(This is the World Wide Node Name (WWNN) of LUN 0)
85
The Target Device Name Descriptor is only returned on SAS devices.

Drive Component Revision Levels Pages (C0h)
Table 36. Drive Component Revision Levels Pages
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Code (C0h)
2 Reserved
3 Page Length (27h)
4
: Code Name (GTYYMMDDsCCC)
15
16
: Reserved (00000000000000h)
22
23
: Date (YYYYMMDD)
30
31
: Reserved (000000000000000000000000h)
42
The Date field contains the date the code was built, with the year defined first.

LOAD/UNLOAD
Table 37. LOAD/UNLOAD Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (1Bh)
1 Obsolete Reserved Immed
2 Reserved
3 Reserved
4 Reserved Hold EOT(0) Reten (0) Load
5 Control
If the Load field is set to 1and there is a tape in the drive, it is positioned to BOM.
If the Load field is set to 1 and there is no tape in the drive, Check Condition
status is returned. The Sense Key is set to Not Ready (2) and the ASC/ASCQ is set
to Parameter Medium Not Present (3A00).
If the Load field is set to 0, there is a tape in the drive, and medium removal
prevention has not been set, then the tape is unloaded and ejected. If the Load
field is set to 0, there is a tape in the drive, and medium removal prevention has
been set, then Check Condition status is returned. The Sense Key is set to Illegal
Request (5) and the ASC/ASCQ is set to Medium Removal Prevented (5302).
If the Load field is set to 0 and the tape is in the ejected position, the command is
presented with Check Condition status and associated sense data of 2/0402 (Not
Ready, Initialization Required). If the Load field is set to 0 and there is no cartridge
present, the command is presented with Check Condition status and associated
data of 2/3A00 (Not Ready, Medium Not Present).
completed.
The Retension (Reten) field is not supported and will be ignored.
The End Of Tape (EOT) field is not supported and should be set to 0. If the EOT
field is set to 1, then Check Condition status is returned. The Sense Key is set to
Illegal Request (5) and the ASC/ASCQ is set to Invalid Field in CDB (2400).
Table Table 38 on page 46 describes the behaviors for the command.

Table 38. LOAD/UNLOAD actions depending on Load and Hold bit settings and medium position
Medium
Position on Behavior and Medium Position at command
Load Hold Statue
command completion
receipt
Check Condition Status
is returned with Not No action is taken
No tape present 0 x
Ready, Medium Not No tape present
Present (2/3A00)
is returned with Not No action is taken
No tape present 1 x
Ready, Medium Not No tape present
Present (2/3A00)
is returned with Not
No action is taken
Ejected Position 0 x Ready, Initializing
Ejected Position
Command Required
(2/0402)
Seated and tape is threaded and ready for access at
Ejected Position 1 0 Good Status returned
LBA 0
Seated with Cartridge Memory (CM) accessible but
Ejected Position 1 1 Good Status returned
tape is not threaded
Seated but tape
0 0 Good Status returned Ejected Position
not threaded
Seated but tape No action taken
0 1 Good Status returned
not threaded Position it was at prior to command
Seated but tape Seated and tape is threaded and ready for access at
not threaded LBA 0
Seated but tape No action taken
not threaded Position it was at prior to command
Seated with tape
0 0 Good Status returned Ejected Position
threaded
Seated with tape Seated with Cartridge Memory (CM) accessible but
threaded tape is not threaded
Seated and tape is rewound and ready for access at
Seated with tape
1 0 Good Status returned LBA 0
threaded
(i.e. this is equal to issuing a rewind command)
Seated with tape is returned with Illegal No action is taken
1 1
threaded Request, Invalid Field Position it was at prior to command
in CDB (5/2400)
The Retension (Reten) bit is not supported and will be ignored.

LOCATE
The LOCATE command causes the logical position on tape to be set to the value
indicated by the Block Address field. The value indicates the total number of
records and marks between BOM and the desired logical position. A value of 0
causes the tape to be positioned at BOM.
Table 39. LOCATE (10) command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved BT(0) CP(0) Immed
2 Reserved
3
: Block Address
6
7 Reserved
8 Partition (0)
9 Control
Table 40. LOCATE (16) Command (Ultrium 4 and Later)

Bit
Byte 7 6 5 4 3 2 1 0
1 Reserved Dest_Type Rsvd CP (0) IMMED
2 Reserved BAM (0)
3 Partition (00h)
4 (MSB)
Block Address
11 (LSB)
12
Reserved
14
15 Control
completed.
If the LOCATE command fails for anything other than Illegal Request, the logical
position is not guaranteed and a READ POSITION command should be issued to
determine the current logical position of the tape.

The Block Type (BT) and Change Partition (CP) fields are not supported and
should be set to 0. The Partition field is not supported and will be set to 0. If the
Partition field is set to anything other than 0, then Check Condition status is
returned. The Sense Key is set to Illegal Request (5) and the ASC/ASCQ is set to
Invalid Field in CDB (2400).
The Dest_Type field is used in conjunction with the Block Address field to locate to
the appropriate position of the medium. The Dest_Type field specifies whether the
location specified is a block address or logical file identifier. The Dest_Type field is
defined in Table 41.
Table 41. Dest_Type field definitions
Code Description Logical Position upon successful completion
00b Block Address BOP side
01b Logical file identified BOP side of the logical file
10b Obsolete
11b Reserved

LOG SELECT
The LOG SELECT command causes log data on the drive to be reset to its default
value or to be set to an initiator-specific value.
Table 42. LOG SELECT Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (4Ch)
1 Obsolete Reserved PCR SP(0)
2 PC Reserved
3 Reserved
4 Reserved
5 Reserved
6 Reserved
7
: Parameter List Length
8
9 Control
If the Parameter Code Reset (PCR) field is set to 1, the Parameter List Length is 0.
The action taken by the drive is specified for the values of the Page Control (PC)
field as follows:
v 00b means that no action is taken and Good status is returned.
v 01b means that all resettable logs on the drive are reset to default values.
v 10b means that no action is taken and Good status is returned.
v 11b means that all resettable logs on the drive are reset to default values.
If the Parameter Code Reset (PCR) field is set to 0, the Parameter List Length is not
0. The action taken by the drive is specified for the values of the Page Control (PC)
field as follows:
v 00b means that Check Condition status is returned. The Sense Key is set to
Illegal Request and the ASC/ASCQ is set to Invalid Field in CDB (2400).
v 01b means that data from the server is written to the indicated logs, provided
that the logs are writable.
v 10b means that Check Condition status is returned. The Sense Key is set to
Illegal Request and the ASC/ASCQ is set to Invalid Field in CDB (2400).
v 11b means that data from the server is written to the indicated logs, provided
that the logs are writable.
The Save Page (SP) field is not supported and must be set to 0.

LOG SENSE
The LOG SENSE command causes log data to be sent to the initiator.
Table 43. LOG SENSE Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (4Dh)
1 Obsolete Reserved PPC(0) SP(0)
2 PC Page Code
3 Reserved
4 Reserved
5
: Parameter Pointer (0)
6
7
: Allocation Length
8
9 Control
The log values returned are controlled by the Page Control (PC) field value as
follows:
v 00b means that the maximum value for each log entry is returned.
Note: For page 2Eh (TapeAlert) only, the PC field is ignored. Current® values
are always returned.
v 01b means that the current values are returned.
v 10b means that the maximum value for each log entry is returned.
v 11b means that the power-on values are returned.
The Parameter Pointer Control (PPC) must be set to 0. Returning changed

parameters is not supported. The Save Page (SP) field must be set to 0. Saved
pages are not supported. The Parameter Pointer will be 0.
Cartridge-specific log parameter counts are set to 0 when a cartridge has

successfully loaded.
WARNING
Log parameter data must be dynamically parsed as some parameters may not
be present and other parameters may be inserted. The relative location of
parameters have changed and are anticipated to continue to change.

Log Page Format
All log pages (except page 0) consist of a log page header, followed by a number
of log parameters. The log page header has the format indicated in Table 44.
Table 44. Log Page Header Format
Bit
Byte 7 6 5 4 3 2 1 0
0 Reserved Page Code
1 Reserved
2
: Page Length
The Page Code is a byte value that uniquely identifies what log page is being
returned. The Page Length describes how many bytes are to follow for the entire
log page.
Each log parameter has the format indicated in Table 45.

Table 45. Log Parameter Format
Bit
Byte 7 6 5 4 3 2 1 0
0
: Parameter Code
1
2 DU DS (1) TSD (0) ETC (0) TMC (0) LBIN LP
3 Parameter Length (n-3)
4
: Parameter Bytes
The Parameter Code is a 2-byte value that uniquely identifies the parameter within
the log.
The Disable Update (DU) field is set for any parameter that the server can neither
write nor reset.
The List Parameter (LP) field is 0 for parameters that are counters and 1 for
parameters that are not counters.
If the LP field is 1 and the parameter is a binary parameter, then the List Binary
(LBIN) field is set to 1. Otherwise it is set to 0.
The TSD, ETC and TMC fields are always 0 and the DS field is always 1.
The Parameter Length field gives the length of the Parameter Bytes field in bytes.
The Parameter Bytes field contains the actual parameter data.

Log Page 00h: Supported Log Pages
The Supported Log Pages Log Page code is 00h. The parameter list contains a
series of 1-byte entries for the log pages that are supported. At least those logs
described in this document are listed. Any additional logs that are supported may
or may not be listed.
WARNING
Table 46. Supported Log Pages Log Page Definition

Product(s) Pointer
Ultrium 1 and Ultrium 2 non-RoHS See Table 47
Ultrium 2 RoHS and Ultrium 3 See Table 48
Ultrium 4 See Table 49 on page 53
Table 47. Supported Log Pages Log Page Format (Ultrium 1 and Ultrium 2 non-RoHS)
Bit
Byte 7 6 5 4 3 2 1 0
0 Reserved Page Code (00h)
1 Reserved
2 (MSB)
Page Length (n-3)
3 (LSB)
4 (00h) Supported Log Pages
5 (02h) Write Error Counters
6 (03h) Read Error Counters
7 (0Ch) Sequential Access Device Log Page
8 (18h) Log Page
9 (2Eh) TapeAlert
10 (30h) Tape Usage Log
11 (31h) Tape Capacity
12 (32h) Data Compression
Table 48. Supported Log Pages Log Page Format (Ultrium 2 RoHS and Ultrium 3)
Bit
Byte 7 6 5 4 3 2 1 0
1 Reserved
2 (MSB)
Page Length (n-3)
3 (LSB)

Table 48. Supported Log Pages Log Page Format (Ultrium 2 RoHS and Ultrium 3) (continued)
Bit
Byte 7 6 5 4 3 2 1 0
7 (06h) Non-Medium Errors
9 (18h) Log Page
10 (2Eh) TapeAlert
14 (33h) Write Errors
15 (34h) Read Forward Errors
16 (38h) Blocks/Bytes Transferred
17 (39h) Host Port 0 Interface Errors
18 (3Bh) Host Port 1 Interface Errors
19 (3Ch) Vendor-Specific
20 (3Dh) Subsystem Errors
Table 49. Supported Log Pages Log Page Format (Ultrium 4)

Bit
Byte 7 6 5 4 3 2 1 0
1 Reserved
2 (MSB)
Page Length (n-3)
3 (LSB)
7 (06h) Non-Medium Errors
9 (18h) Log Page
10 (2Eh) TapeAlert
14 (33h) Write Errors
15 (34h) Read Forward Errors
16 (37h) Performance Characteristics
17 (38h) Blocks/Bytes Transferred
18 (39h) Host Port 0 Interface Errors
19 (3Bh) Host Port 1 Interface Errors

Table 49. Supported Log Pages Log Page Format (Ultrium 4) (continued)
Bit
Byte 7 6 5 4 3 2 1 0
20 (3Ch) Vendor-Specific
21 (3Dh) Subsystem Errors
This data can be neither reset nor written.

Log Page 02h: Write Error Counters
The Write Error Counters log is page 02h. Parameters 0 through 2 are not
supported and are returned as 0. All parameter lengths are 4 bytes long, except
parameter 8000 which is 8 bytes long.
WARNING
Table 50. Write Error Log Parameters

Parameter Code
(in Hex) Counter: Description Size

0000h Not supported 4
0003h Total Corrected Write Errors: These errors are corrected by ECC and do not 4
require error recovery procedures (ERPs). Each count represents one block in error
that was corrected and written.
0004h Total Write Retries: 4
0005h Total Write Kilobytes Processed: Each count represents a kilobyte (1024 bytes) of 4
data processed across the host interface during write-type commands. The count
does not include ERP retries.
0006h Total Uncorrected Write Errors: The total number of write errors that could not be 4
corrected by ECC, no servo error was reported, and the error was not a transient
error. Each count represents one block in error that was not corrected, but was
recovered by ERPs and successfully written.
8000h Unspecified 8
8001h Unspecified 4
This data can be reset to 0, but cannot be written.

Log Page 03h: Read Error Counters
The Read Error Counters log is page 03h. Parameters 0 through 2 are not
supported and are returned as 0. All parameter lengths are 4 bytes long, except
parameter 8000 which is 8 bytes long.
WARNING
The supported fields are listed in Table 51.

Table 51. Read Error Log Parameters
Parameter Code
(in Hex) Counter: Description Size

Total Corrective Read Errors:
0003h These are errors that are corrected by ECC and do not require error recovery 4
procedures (ERPs). Each count represents one block in error that was corrected
and read.
0004h Total Read Retries: 4
Total Read Kilobytes Processed: Each count represents a kilobyte (1024 bytes)
0005h processed across the host interface during read-type commands. The count 4
does not include ERP retries.
Total Uncorrected Read Errors: The total number of read errors that could not
be corrected by ECC, no servo error was reported, and the error was not a
0006h 4
transient error. Each count represents one block in error that was not corrected,
but was recovered by ERPs and successfully read.
8000h Unspecified 8
This data can be reset to 0, but not written.

Log Page 06h: Non-Medium Errors (Not Ultrium 1 or Ultrium 2
non-RoHS)
This page sums the occurrences of error events other than write or read failures.
parameter codes do not discriminate among the various types of events.
WARNING
Table 52. Non-Medium Errors log parameter codes

Parameter Code
Counter: Description Size
(in Hex)
0000h Total Non-Medium Error Count: 4

Log Page 0Ch: Sequential Access Device Log
The Sequential Access Device Log Page is 0Ch.
WARNING
Table 53. Sequential Access Device Log Parameters

Parameter Code
Counter: Description Length
(in Hex)
Number of data bytes received from an initiator(s) during WRITE
0000h 8
command operations.
Number of data bytes written to the media as a result of WRITE
0001h command operations, not counting ECC and formatting overhead. This is 8
also the number of data bytes after compression.
Number of data bytes read from the media during READ command
0002h operations, not counting ECC and formatting overhead. This is also the 8
number of compressed data bytes read from media before decompression.
Number of data bytes transferred to the initiator(s) during READ
0003h 8
command operations.
0100h Cleaning Required 8
Megabytes processed to tape since last cleaning (written after
8000h 4
compression/read before decompression)
8001h Lifetime load cycles of the drive 4
8002h Lifetime cleaning cycles of the drive 4
8003h Lifetime Power-on time (in seconds) 4
A non-zero value of the Cleaning Required parameter indicates that a condition

requiring cleaning has been detected and a subsequent cleaning cycle has not been
completed. The Cleaning Required parameter is persistent across hard resets and
power cycles.

Log Page 2Eh: Tape Alert
The TapeAlert log page is page 2Eh. There are 64 parameters, numbered from 1
through 64 (01h through 40h). Table 54 shows the parameters that are supported
for all generations of the Ultrium Tape Drive. The supported parameters are 0 in
the absence of the condition that generates the flag and are set to a non-zero value
when the condition occurs. All unsupported parameters are always set to 0.
All parameters are 1 byte long. Each parameter is either 0 to indicate that the
corresponding condition has not occurred or non-zero to indicate that the
corresponding condition has occurred. All log parameters are set to 0 when the log
is read. The Log parameters are also set to 0 at power-on, on a reset condition, or
by a LOG SELECT command. Specific flags may be set to 0 when corrective action
has removed the condition that caused the flag to be set. For all parameters, the
DU field is 1, the LP field is 0, and the LBIN field is 0.
The PC field for this page is ignored. Current values are always returned.
For a description of service actions associated with the supported parameters, refer
to the IBM TotalStorage LTO Ultrium 2 Tape Drive Models T400 and T400F Setup,
Operator, and Service Guide, the IBM Ultrium Internal Tape Drive Models T200 and
T200F Setup, Operator, and Service Guide, or the IBM 3580 Ultrium Tape Drive Setup,
Operator, and Service Guide.
WARNING
Table 54. TapeAlert Log Parameters

Parameter Code
In Hex In Decimal Description Set Clear1 Type
01h - 02h 1-2 N/A
The operation has stopped because an error
03h 3 has occurred while reading or writing data R Warning
which the drive cannot correct.
Media can no longer be written or read, or
04h 4 R Critical
performance is severely degraded.
05h 5 The tape is damaged or the drive is faulty. R Critical
The drive can no longer write data to the
06h 6 R Critical
tape.
07h 7 Media Life E L Warning
08h 8 The cartridge is not data-grade. R Warning
The WRITE command was attempted to a
09h 9 R Critical
write-protected tape.
A manual or software unload was attempted
0Ah 10 R Informational
when Prevent Media Removal was on.
0Bh 11 The tape in the drive is a cleaning cartridge. R Informational
You attempted to load a cartridge with an
0Ch 12 unsupported tape format (for example, R Informational
Ultrium 2 cartridge in Ultrium 1 drive).
0Dh - 0Eh 13 - 14 N/A
0Fh 15 The memory chip failed in the cartridge. R Critical

Table 54. TapeAlert Log Parameters (continued)
The operation has failed because the tape
10h 16 cartridge was manually ejected while the L Warning
tape drive was actively writing or reading.
11h 17 Media loaded is Read-Only format. E R Informational
The tape directory on the tape cartridge has
been corrupted. File search performance will
12h 18 be degraded. The tape directory can be R Warning
rebuilt by reading all the data on the
cartridge.
13h 19 Nearing media life R Informational
14h 20 Clean now C Critical
15h 21 Clean periodic C Warning
16h 22 Expired cleaning media C L Critical
17h 23 Invalid cleaning tape C R Critical
18h - 1Dh 24 - 29 N/A
The drive has a hardware fault that requires
1Eh 30 Critical
a reset to recover.
The drive has a hardware fault that is not
1Fh 31 related to the read/write operation, or the Critical
drive requires a power cycle to recover.
20h 32 The drive has identified an interface fault. Warning
21h 33 Eject media U, R Critical
22h 34 Firmware download failed Warning
23h 35 N/A
24h 36 The drive’s temperature limits are exceeded. S Warning
25h 37 The drive’s voltage limits are exceeded. S Warning
26h 38 Predictive failure of drive hardware Critical
27h 39 Diagnostics required Warning
28h - 32h 40 - 50 N/A
33h 51 Tape directory invalid at unload E L Warning
34h 52 Tape system area write failure E R Critical
35h 53 Tape system area read failure E R Critical
36h 54 N/A
37h 55 Loading failure E R Critical
38h 56 Unrecoverable unload failure E U Critical
39h-3Ah 57-58 N/A
(WORM Medium - integrity check failed) set
3Bh 59 when the drive determines that the data on L Critical
tape is suspect from a WORM point of view.
(WORM Medium - Overwrite Attempted) set
when the drive rejects a Write operation
3Ch 60 E Critical
because the rules for allowing WORM writes
have not been met.)
3Dh-40h 61 - 64 N/A

Table 54. TapeAlert Log Parameters (continued)
Legend:
TapeAlert Set/Cleared when:

N/A Not set/supported
L Load - medium is loaded
C Clean - cleaner tape is loaded
U Unload - medium is ejected
E Error - error code is posted
R Removal - medium is FULLY removed
S Sensor - sensor check
Note:
1
All TapeAlerts are cleared on POR/Reset.

Log Page 11h: DT Device Status Log Page
The DT Device Status log page (see Table 55) defines log information pertaining to
the DT device (i.e. tape drive) and DT device primary ports.
WARNING
Table 55. DT Device Status Log Page

Bit
Byte 7 6 5 4 3 2 1 0
1 Reserved
2 (MSB)
Page Length (n-3)
3 (LSB)
4
DT Device Status Log Parameters
n
Table 56 defines the DT Device Status log page parameter codes.

Table 56. DT Device Status log page parameter codes
Parameter Code (in Hex) Description Reference
0000h Very high frequency data See Table 57.
0001h Very high frequency polling See Table 57.
delay
0002h - 00FFh Reserved See Table 57.
0100h - 0200h DT device primary port See Table 61 on page 66.
status
Very High Frequency Data Log Parameter

The very high frequency data log parameter format is shown in Table 57.
Table 57. Very high frequency data log parameter format
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
PARAMETER CODE (0000h)
1 (LSB)
2 DU (0) DS (1) TSD (0) ETC (0) TMC (0) LBIN (1) LP (1)
3 PARAMETER LENGTH (04h)
4
VHF DATA DESCRIPTOR
7
The PARAMETER CODE field is set to 0000h to indicate the very high frequency
data log parameter.

The PARAMETER LENGTH field is set to 04h to allow transfer of the complete
parameter.
The VHF DATA DESCRIPTOR field is defined in Table 58. Returned data shall
reflect the last known values since the DT device initialized.
Table 58. VHF DATA DESCRIPTOR field
Bit
Byte 7 6 5 4 3 2 1 0
4 PAMR HIU MACC CMPR WRTP CQRST CRQRD DINIT
5 MTHRD MTHRD
INXTN Rsvd RAA® MPRSNT Rsvd MSTD
MOUNTED MOUNTED
6 DT DEVICE ACTIVITY
7 VS Reserved RRQST INTFC TAFC
Note: In addition to reliance on indication of initialization completion, reliance on

returned values should also take into consideration conditions indicated by
changes in Tape Alert flag status, and process those first as needed.
A DT device initialized (DINIT) bit set to one indicates that the DT device is able
to return valid very high frequency data. A DINIT bit set to zero indicates DT
device initialization is required or incomplete. The DINIT bit should be set to one
before relying on any other bits in the very high frequency data log parameter.
A cleaning required (CRQRD) bit set to one indicates that a head cleaning
operation is required before a data medium is able to reach load state (i) (see
ADC-2), and that normal operation may not be possible if the head cleaning
operation is not performed. A CRQRD bit set to zero indicates that urgent cleaning
is not required. The CRQRD bit shall take priority over the CRQST bit. It shall not
be considered an error for the CRQRD bit and the CRQST bit to both be set to one.
A cleaning requested (CRQST) bit set to one indicates that the DT device has
requested a head cleaning. A CRQST bit set to zero indicates that no cleaning is
requested.
A write protect (WRTP) bit set to one indicates that any currently present medium
is physically write protected. A WRTP bit set to zero indicates that any currently
present medium is not physically write protected. The WRTP bit is only valid if the
MPRSNT bit is set to one. The WRTP bit should be set to zero if the MPRSNT bit
is set to zero.
Note: Physically write protected refers to any mechanism used within the medium
shell itself to write protect the medium (for example, sliding windows or
tabs) and not logical states of write protection caused by commands to the
DT device.
A compress (CMPR) bit set to one indicates that the DT device currently has data
compression enabled. A CMPR bit set to zero indicates that compression is not
enabled.
A medium auxiliary memory accessible (MACC) bit set to one indicates that the
medium is located at a position where the Medium Auxiliary Memory (MAM) is
accessible. A MACC bit set to zero indicates that the MAM is not accessible. If the
MACC bit is set to one, the ADC device server shall also support commands to
access the MAM. If the MACC bit is supported the MACC bit should only be set
to one if the MPRSNT bit is set to one. The MACC bit is only applicable for drives
and media that support MAM.
The host initiated unload (HIU) bit is set to one when the drive reaches any one of
the unload states (e) - (h) (see ADC-2r07e, Clause 4.4.2), due to the RMC device
server receiving a LOAD UNLOAD command (see SSC-2) with the LOAD bit set
to zero. The HIU bit is set to zero when the drive transitions to any state in table 2
or table 4 of ADC-2r07e other than unload states (e) - (h) in table 4 of ADC-2r07e.
The HIU bit may be set to zero following a logical unit reset of the RMC or ADC
device servers.
The prevent/allow medium removal (PAMR) bit is set to one when removal of the
medium in the DT device is prevented as the result of the RMC device server
processing a PREVENT/ALLOW MEDIUM REMOVAL command (see SPC-3 or the
relevant command set standard).
The PAMR bit is set to zero when removal of the medium in the DT device is
allowed as defined by the PREVENT/ALLOW MEDIUM REMOVAL command.
A MOUNTED bit set to one indicates that the DT device is in load state (i) (see
ADC-2). The MOUNTED bit set to one may correspond to the RMC device server
being able to respond to a TEST UNIT READY command with a status of GOOD,
however when a cleaning or microcode image medium is loaded the RMC device
server may respond to a TEST UNIT READY command with a CHECK
CONDITION with the sense key set to NOT READY. A MOUNTED bit set to zero
indicates that the DT device is not in load state (i).
A medium threaded (MTHRD) bit set to one indicates that the medium has been
threaded by the DT device, such that tape motion operations are possible. A
MTHRD bit set to zero indicates that the medium has not been threaded.
Note: The value of the MTHRD bit may or may not correspond to the DT device
responding with a status of GOOD to a TEST UNIT READY command, as
additional processing may be required by the DT device after threading
before the logical unit becomes ready.
A medium seated (MSTD) bit set to one indicates that the medium is mechanically
seated within the loading mechanism (i.e., the physical loading process has
completed). A MSTD bit set to zero indicates that the medium is not seated, and
that further mechanical motion remains in order to complete the loading process,
exclusive of tape threading.
A medium present (MPRSNT) bit set to one indicates that the DT device detects
the presence of a medium. A MPRSNT bit set to zero indicates that the DT device
does not detect a medium present.
A robotic access allowed (RAA) bit set to one indicates that the automation device
may move a medium to or from the DT device. A RAA bit set to zero indicates
that the automation device should not move a medium to or from the DT device.
The DT device should indicate that access is allowed by the robotics if a medium
may be successfully inserted into or removed from the DT device.
Note: The RAA bit is not intended to reflect the value of any PREVENT/ALLOW
MEDIUM REMOVAL command settings (see SPC-3), nor the ability of the
automation device to issue commands to the DT device.

The in transition (INXTN) bit governs the remaining bits within byte 1 to indicate
the stability of the values returned and whether state transitions are taking place.
An INXTN bit set to one indicates that the state currently reflected by the
remaining bits in byte 1 is in transition, because the DT device is transitioning to
another state. An INXTN bit set to zero indicates that the DT device is in the state
reflected by the remaining bits in byte 1 and is making no attempt to leave this
state. When the recovery requested (RRQST) bit is set to one, the INXTN bit is set
to zero.
The DT DEVICE ACTIVITY field is used to describe the current activity of the DT
device (see Table 59).
Table 59. DT DEVICE ACTIVITY field values
Value Description
00h No DT device activity
01h Cleaning operation in progress
02h Medium is being loaded
03h Medium is being unloaded
04h Other medium activity
05h Reading from medium
06h Writing to medium
07h Locating medium
08h Rewinding medium
09h Erasing medium
0Ah Formatting medium
0Bh Calibrating medium
0Ch Other DT device activity
0Dh Microcode update in progress
A TapeAlert state flag changed (TAFC) bit set to one indicates that at least one
TapeAlert state flag has changed from its previous value since the last retrieval of
the TapeAlert Response log page by this I_T nexus.
The ADC device server sets the TAFC bit to zero after retrieval of the TapeAlert
Response log page by this I_T nexus. A TAFC bit set to zero indicates that no
TapeAlert state flag has changed. There may not be any difference in the TapeAlert
state flags upon retrieval if the state changed again between the time of reporting
through the TAFC bit and retrieving the TapeAlert Response log page. This should
not be considered an error. The TAFC bit should be processed following the DINIT
bit. Pending TapeAlert state flags may affect the reliability of the values returned in
other bits within the VHF DATA DESCRIPTOR.
An interface changed (INTFC) bit set to one indicates that one or more fields in the
DT device primary port status log parameters have changed since the last retrieval
of any of the DT device primary port status log parameters from the DT Device
Status log page by this I_T nexus. An INTFC bit set to zero indicates that one or
more fields in the DT Device Primary Port Status log parameters have not changed
since the last retrieval of any of the DT device primary port status log parameters

by this I_T nexus. The INTFC bit is set to zero after retrieval of any of the DT
device primary port status log parameters from the DT Device Status log page by
this I_T nexus.
The recovery requested (RRQST) bit is set to one to indicate that the DT device has
detected an error and that one or more requested recovery procedures are available
via the ADC Requested Recovery log page. A RRQST bit set to zero indicates that
no recovery procedure is requested. The RRQST bit shall remain set to one as long
as a recovery procedure is available. When the RRQST bit is set to one, the INXTN
bit is set to zero.
Note: The Requested Recovery log page may indicate that a recovery procedure is
not requested or not defined.
Very High Frequency Polling Delay Log Parameter

The very high frequency polling delay log parameter format is shown in Table 60.
Table 60. Very high frequency polling delay log parameter format
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Parameter Code (0001h)
1 (LSB)
3 PARAMETER LENGTH (02h)
4 (MSB)
VHF POLLING DELAY
5 (LSB)
The PARAMETER CODE field is set to 0001h to indicate the very high frequency
polling delay log parameter.
The PARAMETER LENGTH field is set to 02h to allow transfer of the complete
parameter.
The VHF POLLING DELAY field indicates the minimum delay in milliseconds the
automation device should wait before requesting another DT Device Status log
page.
Primary Port Status Descriptor

The Primary Port Status descriptor is defined in Table 61. This descriptor reports
the current operating points of the specified port.
Table 61. Primary Port Status descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Parameter Code
1 (LSB)
3 PARAMETER LENGTH (8)
4 CURRTOP CURRENT SPEED LC CONFLICT SIGNAL PIC

Table 61. Primary Port Status descriptor (continued)
Bit
Byte 7 6 5 4 3 2 1 0
5
6 Current N_Port_ID
7
8 Reserved
9 Reserved
10 Reserved
11 Rsvd CURRENT FC-AL LOOP ID
The Parameter Code field contains a value from 0101h to 01FFh which uniquely
identifies the primary port relative to other primary ports in the drive. Once
assigned, the Parameter Code value for a port will not be changed. For each
primary port, the Parameter Code value is equal to 0100h plus the value of the
Relative Target Port field associated with that port. The Relative Target Port field
contains a value assigned by the device that uniquely identifies the port relative to
other ports in the device, independent of port type. The relative target port value
for a port will not be changed. The Relative Target Port is the same as the relative
target port value defined in the VPD pages of inquiry. A value of 1=port 0 or port
A; a value of 2=port 1 or port B. All values in this descriptor relates to the primary
port specified by the Parameter Code.
The Additional Primary Port Status Length field specifies the additional bytes in
the current Primary Port Status descriptor. Currently this is 4, but this is subject to
change should the ADC specification for the DTD Primary Port Status log
parameters change
A port initialization complete (PIC) bit set to one indicates that the FC_Port state
machine is in the ACTIVE state (see FC-FS) and the drive’s primary port is
operating in point-to-point topology, or the most recent Loop Initialization Process
(LIP) has completed successfully (see FC-AL-2). A PIC bit set to zero indicates that
the drive’s primary port is not in the ACTIVE state and is not synchronized (see
FC-FS), or has not successfully completed the most recent LIP.
A value of one in the SIGNAL field indicates that a signal is detected at the
primary port. A value of zero indicates a signal is not detected. An example of
signal detection is detection of light for an optical medium.
A value of one in the CONFLICT field indicates that another device has the
required Hard AL_PA or that no AL_PA is available for the primary port. A value
of zero indicates there is no AL_PA conflict.
A login complete (LC) bit set to one indicates that at least one host is currently
logged in to the drive through this port. Logged in is defined as having
successfully completed a process login (PRLI) with the drive (see FCP-2) and still
have an active session. A host is defined as an initiator port that does not have a
Source ID (S_ID) that is a Well-Known name (see FC-FS). An LC bit set to zero
indicates that a host is not currently logged in on this port.

The SPEED field indicates the bit rate that the port is configured to operate in. The
valid values can be found in Table 62. This field is undefined when the PIC field is
set to zero.
Table 62. Speed Values
Value Speed
000b 1 Gb/sec.
001b 2 Gb/sec.
010b 4 Gb/sec.
100b - 111b Reserved
A value of one in the current topology (CURRTOP) field indicates the port is
operating currently in point to point mode. A value of zero indicates the port is
operating currently in arbitrated loop mode. This field is undefined when the PIC
field is set to zero.
The Current N_Port_ID field indicates the 24-bit N_Port_ID (as defined by FC-FS)
that is assigned currently to the port. This field is undefined when the PIC field is
set to zero.
The CURRENT FC-AL LOOP ID field indicates the loop identifier (see FC-AL-2)
that is assigned to the drive’s primary port associated with the PARAMETER
CODE. The CURRENT FC-AL LOOP ID field is ignored when the PIC bit is set to
zero or when the CURRTOP bit is set to one.

Log Page 14h: Device Statistics Log Page (Not Ultrium 1 or 2)
The Device Statistics log page defines data counters associated with utilization of
the tape drive.
All supported parameters are persistent across I_T nexus loss, logical unit reset
and power-on. The parameters are not set to zero or changed with the use of a
LOG SELECT command.
WARNING
Table 63. Device Statistics log parameter codes

Parameter Code (in
Counter: Description Size (decimal)
Hex)
0000h Lifetime media loads 4
0001h Lifetime cleaning operations 4
0002h Lifetime power on hours 4
0003h Lifetime media motion (i.e., head) hours 4
0004h Lifetime meters of tape processed 4
Lifetime media motion (head) hours when incompatible media was last
0005h 4
loaded
Lifetime power on hours when the last temperature condition occurred
0006h 4
(i.e., TapeAlert code 24h)
Lifetime power on hours when the last power consumption condition
0007h 4
occurred (i.e., TapeAlert code 1Ch)
0008h Media motion (i.e., head) hours since last successful cleaning operation 4
Media motion (i.e., head) hours since second to last successful cleaning
0009h 4
operation
Media motion (i.e., head) hours since third to last successful cleaning
000Ah 4
operation
Lifetime power on hours when the last operator initiated forced reset
000Bh 4
and/or emergency eject occurred
000Ch-0FFFh Reserved
1000h Media motion (i.e., head) hours for each medium type (See Table 64) 8*n
n = Number of Medium Types supported (see Table 64 Media Type Values for number supported by each
generation)
The Medium Type Log Parameter format is specified in Table 64.

Table 64. Medium Type Log Parameter format
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
1 (LSB)

Table 64. Medium Type Log Parameter format (continued)
Bit
Byte 7 6 5 4 3 2 1 0
2 FORMAT AND
DU(0) Obsolete TSD(0) ETC(0) TMC (00b)
LINKING (11 b)
Medium type parameter(s)
4 Medium type log parameter (first) (see Table 65)
.
.
.
n Medium type log parameter (last) (see Table 65)
Table 65. Medium Type log parameter

Bit
Byte 7 6 5 4 3 2 1 0
0 Reserved
1 Reserved
2 Density Code
3 Medium Type
4 (MSB)
Media Motion Hours
7 (LSB)
The Density Code field contains the value returned in the general mode parameter
block descriptor.
The Medium Type field contains the value returned in the mode parameter header
(see Mode Parameter Header).
The Media Motion Hours field contains the number of media motion (i.e., head)
hours for the type of medium specified by the combination of the Medium Type
field and Density Code field.
The Medium Type log parameter is specified in Table 65. One parameter exists for
each Medium Type/Density Code combination that has been loaded in the drive.

Log Page 16h: Tape Diagnostic Data Log Page (Not Ultrium 1
or 2)
The Tape Diagnostic Data log page (see Table 66) provides for a number of
error-event records using the list parameter format. Each error-event record
contains diagnostic information for a single error type encountered by the device
including data counters associated with the error event, sense data, operation
code/service action and medium type with associated media motion hours, etc.
The Tape Diagnostic Data log page may be used to aid in field analysis and repair.
The Tape Diagnostic Data log page only includes parameter entries for commands
that terminated with a CHECK CONDITION status having the sense key set to
MEDIUM ERROR, HARDWARE ERROR or ABORTED COMMAND.
The parameter code value associated with an error-event indicates the relative time
at which a command terminated with a CHECK CONDITION status. A lower
parameter code indicates that the command terminated with a CHECK
CONDITION status at a more recent time. The parameter code values returned is
numbered consecutively from 0000h (i.e., the most recent) up to n, where n is the
number of current parameter entries.
In each parameter (see Table 67 on page 72) if the REPEAT bit is set to zero, then
the parameter represents only one event. If the REPEAT bit is set to one, then the
parameter represents more than one consecutive events that had identical values
for the MEDIUM ID NUMBER field, SENSE KEY field, ADDITIONAL SENSE
CODE field and ADDITIONAL SENSE CODE QUALIFIER field in the parameter.
If the REPEAT bit is set to one in the parameter, then other fields in the parameter
is set to the values when the first of the consecutive events that had the identical
values for the MEDIUM ID NUMBER field, SENSE KEY field, ADDITIONAL
SENSE CODE field and ADDITIONAL SENSE CODE QUALIFIER field occurred.
All parameter codes are persistent across I_T nexus losses, logical unit resets, and
power-on. The parameter entries are not set to zero or changed with the use of a
LOG SELECT command.
WARNING
Table 66. Tape Diagnostic Data log page

Bit
Byte 7 6 5 4 3 2 1 0
0 DS SPF(0) Page Code (16h)
1 Subpage Code (00h)
2 (MSB)
Page Length (n-3)
3 (LSB)
Tape Diagnostic Data log parameter(s)
4 Tape Diagnostic Data log parameter (first)
n Tape Diagnostic Data log parameter (last)

The Tape Diagnostic Data log parameter format is specified in Table 67.
Table 67. Tape Diagnostic Data log parameter format
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
1 (LSB)
FORMAT AND
2 DU(0) Obsolete TSD(0) ETC(0) TMC (00b)
LINKING (11 b)
4 Reserved
5 Reserved
6 Density Code
7 Medium Type
8 (MSB)
Lifetime Media Motion Hours
11 (LSB)
12 Reserved
13 Repeat Reserved Sense Key
14 Additional Sense Code
15 Additional Sense Code Qualifier
16 (MSB)
Vendor-Specific Code Qualifier
19 (LSB)
20 (MSB)
Product Revision Level
23 (LSB)
24 (MSB)
Hours Since Last Clean
27 (LSB)
28 Operation Code
29 Reserved Service Action
30 Reserved
31 Reserved
32 (MSB)
Medium Id Number
63 (LSB)
64 Reserved Timestamp Origin
65 Reserved
66 (MSB)
Timestamp
71 (LSB)
The PARAMETER LENGTH field indicates the number of bytes in the Tape
Diagnostic Data log parameter data that follows.
The DENSITY CODE field contains the density code of the medium loaded at the
time the command terminated with the CHECK CONDITION status. The
DENSITY CODE field is the same value as returned in the general mode parameter
block descriptor. If no medium was loaded at the time the command terminated
with the CHECK CONDITION status, then the DENSITY CODE field is set to 00h.

The MEDIUM TYPE field contains the type of medium loaded at the time the
command terminated with the CHECK CONDITION status. The MEDIUM TYPE
field is the same value as returned in the mode parameter header. If no medium
was loaded at the time the command terminated with the CHECK CONDITION
status, then the MEDIUM TYPE field is set to 00h.
The LIFETIME MEDIA MOTION HOURS field contains the number of media
motion (head) hours at the time the command terminated with the CHECK
CONDITION status. The LIFETIME MEDIA MOTION HOURS field is equivalent
to the value contained in the Device Statistics log page with a parameter code
value of 0003h at the time the command terminated with the CHECK CONDITION
status.
The REPEAT bit set to one indicates this parameter represents more than one
consecutive events that had identical values for the MEDIUM ID NUMBER field,
SENSE KEY field, ADDITIONAL SENSE CODE field, and ADDITIONAL SENSE
CODE QUALIFIER field. The REPEAT bit set to zero indicates this parameter
represents a single event.
The SENSE KEY field, ADDITIONAL SENSE CODE field, and ADDITIONAL
SENSE CODE QUALIFIER field contain the sense key and additional sense code
values of the command that terminated with the CHECK CONDITION status.
The VENDOR-SPECIFIC CODE QUALIFIER field contains the Last Error FSC and
Last Error Flag Data (i.e. bytes 65-68 of Sense Data) generated for the command
that terminated with the CHECK CONDITION status.
The PRODUCT REVISION LEVEL field contains the product revision level at the
time the command terminated with the CHECK CONDITION status.
The HOURS SINCE LAST CLEAN field contains the time in media motion (i.e.,
head) hours since the last successful cleaning at the time the command terminated
with the CHECK CONDITION status. The HOURS SINCE LAST CLEAN field is
equivalent to the value contained in the Device Statistics log page with a
parameter code of 0008h at the time the command terminated with the CHECK
CONDITION status.
The OPERATION CODE field and SERVICE ACTION field if applicable contain
the operation code and service action of the command that terminated with the
CHECK CONDITION status.
If medium was present at the time the command terminated with the CHECK
CONDITION status, then the MEDIUM ID NUMBER field contains (in prioritized
order):
1. the BARCODE field value contained in the medium auxiliary memory;
2. the MEDIUM SERIAL NUMBER field value contained in the medium auxiliary
memory; or
3. the VOLUME IDENTIFIER field value contained in the medium auxiliary
memory.
The TIMESTAMP ORIGIN field and TIMESTAMP field contain the timestamp
origin and timestamp maintained by the tape drive at the time the command
terminated with the CHECK CONDITION status. If a timestamp is not supported
by the device server, the TIMESTAMP ORIGIN and TIMESTAMP fields is set to
zero.

If no medium was present at the time the command terminated with the CHECK
CONDITION status, the MEDIUM ID NUMBER field is filled with 20h (i.e., ASCII
space).
Log Page 18h: Protocol-Specific Logical Unit Log Page

There is one copy of this page for each initiator. This page is defined for
SAS-attached devices only.
Support Protocol-Specific log parameter for SAS as in spec. (attached means other
end of cable from drive - initiator or expander).
The Protocol-Specific log page for SAS defined in Table 68 is used to report errors
that have occurred on the phys of SAS drives.
WARNING:
This page must be dynamically parsed as some parameters may not be present and
other parameters may be inserted. The relative location of parameters have
changed and are anticipated to continue to change.
Table 68. Protocol-Specific log page for SAS
Byte/Bit 7 6 5 4 3 2 1 0
0 Page Code (18h)
1 Reserved
2 (MSB)
Page Length (m-3)
3 (LSB)
Protocol-specific log parameter list
4
Protocol-specific log parameter (first) (See Table 69 on page 75)
.
.
.
Protocol-specific log parameter (last) (See Table 69 on page 75)

m
Table 69 on page 75 defines the format for the Protocol-Specific log parameter for
SAS.

Table 69. Protocol-Specific log parameter for SAS
Byte/Bit 7 6 5 4 3 2 1 0
0 (MSB)
Parameter Code (relative target port identifier)
1 (LSB)
2 Parameter control byte
DU (0) DS (0) TSD (0) ETC (0) TMC (0) LBIN (1) LP (1)
3 Parameter Length (y-3)
4 Reserved Protocol Identifier (6h)
5
Reserved
6
7 Number of PHYs (01h)
SAS phy log descriptor list
8
SAS phy log descriptor (first) (See Table 70 on page 76)
.
.
.
SAS phy log descriptor (last) (See Table 70 on page 76)

y
Table 70 on page 76 describes the SAS Phy log descriptor.

Table 70. SAS Phy log descriptor
Byte/Bit 7 6 5 4 3 2 1 0
0 Reserved
1 Phy Identifier
2
Reserved
3
4 Rsvd Attached Device Type Reserved
5 Reserved Negotiated Physical Link Rate
6 Attached Attached Attached Rsvd
Reserved SSP STP SMP
Initiator Initiator Initiator
7 Attached Attached Attached Rsvd
Reserved SSP STP SMP
Target Target Target
8
SAS Address
15
16
Attached SAS Address
23
24 Attached Phy Identifier
25
Reserved
31
32 MSB
Invalid Dword Count
35 LSB
36 MSB
Running Disparity Error Count
39 LSB
40 MSB
Loss Of Dword Synchronization
43 LSB
44 MSB
Phy Reset Problem
47 LSB

Log Page 30h: Tape Usage Log
The Tape Usage Log Page Code is 30h. These are all read directly from the tape
logs.
WARNING
Table 71. Tape Usage Log Parameters

Parameter Code
Counter: Description
(in Hex) Size
Thread Count
0001h 4
This field is incremented by one each time the cartridge has been loaded.
Total Data Sets Written
0002h 8
Total number of Data Sets written to the tape.
Total Write Retries
U13: If an error is detected in a CQ Set during Read-While-Write verification,
then the Total Write Retries counter shall increment by one regardless of the
number of times that that Data Set is retried.
0003h 4
U4: If an error is detected in a CQ Set during Read-While-Write verification, then
that CQ Set is rewritten farther down the tape. If the Data Set was successfully
rewritten, then the Total Write Retries counter shall increment by one regardless
of the number of times that that Data Set is retried.
Total Unrecovered Write Errors
If an error is detected in a CQ Set during Read-While-Write verification, then
0004h that CQ Set is rewritten farther down the tape. If the Data Set cannot be 2
successfully written, then the Total Unrecovered Write Errors counter shall
increment by one.
Total Suspended Writes
U13: If a defect or disturbance that may result in incorrectly written tracks is
detected while recording data, recording shall stop and this parameter is
incremented. If the Data Set is interrupted one or more times, this parameter is
incremented for the first interruption only.
0005h 2
U4: When writing is interrupted by a defect or disturbance that results in
incorrectly written tracks is detected while recording data, and has to write the
Data Set further down the tape, this parameter is incremented. If the Data Set is
interrupted one or more times, this parameter is incremented for the first
interruption only.
Total Fatal Suspended Writes
U13: If a defect or disturbance that may result in incorrectly written tracks is
detected while recording data, recording shall stop. If the Data Set cannot be
successfully written, then this parameter shall increment by one.
0006h 2
U4: When writing is interrupted by a defect or disturbance that results in
incorrectly written tracks is detected while recording data, and has to write the
Data Set further down the tape but cannot be successfully written, this
parameter is incremented.
Total Data Sets Read
0007h 8
Total number of Data Sets read from to the tape.

Table 71. Tape Usage Log Parameters (continued)
Parameter Code
Counter: Description
(in Hex) Size
Total Read Retries
U13: Total number of Data Sets that were not successfully read on the first
0008h attempt by a dataflow error. 4
attempt, but which were subsequently successfully read.
Total Unrecovered Read Errors
U13: Total number of Data Sets that were not successfully read on the first, nor
0009h on a subsequent attempt to read it by a dataflow error. 2
attempt, but which were subsequently successfully read.
Total Suspended Reads
U13: If a defect or disturbance that may result in incorrectly read tracks is
detected while reading data, reading shall stop and this parameter is
000Ah 2
incremented. If the Data Set is interrupted one or more times, this parameter is
incremented for the first interruption only.
U4: (0000h) This parameter is never updated.
Total Fatal Suspended Reads
U13: If a defect or disturbance that may result in incorrectly read tracks is
000Bh detected while reading data, reading shall stop. If the Data Set cannot be 2
successfully read, then this parameter shall increment by one.
U4: (0000h) This parameter is never updated.
U13: Ultrium 1, Ultrium2, and Ultrium 3
U4: Ultrium 4

Log Page 31h: Tape Capacity Log
The Tape Capacity Log Page Code is 31h. Parameters 2 and 4 are not supported
and are returned as 0. All parameter lengths are 4 bytes long. The supported fields
are listed in Table 72.
WARNING
Table 72. Tape Capacity Log Parameters

Parameter Code
(in Hex)
Main Partition Remaining Capacity: A representation of the location on
medium of EOD expressed in Native Capacity between EOD and EOP. This is
0001h intended to be used as a gauge to display the percentage of medium used. 4
There is no guarantee that the amount of data expressed in this parameter will
be available for writing.
Alternate Partition Remaining Capacity: (Not supported - Always returns
0002h 4
00000000h)
Main Partition Maximum Capacity: (Constant value representing the Native
Capacity of the medium)
0003h Ultrium 1: 0001 7487 4

Ultrium 2: 0002 E90E
Ultrium 3: 0005 D21D
Ultrium 4: 000B A43B
Alternate Partition Maximum Capacity: (Not supported - Always returns
0004h 4
00000000h)
All parameters are in megabytes and assume no data compression. This data
cannot be reset or written.
Note: For this command, a megabyte is equal to 1 048 576 bytes. As an example, a
value of 17487h in Parameter 3 is equal to 95 367 megabytes, which is equal
to 100 000 000 000 bytes.

Log Page 32h: Data Compression Log
The Data Compression Log Page Code is 32h. Parameter byte fields 0 and 1 are 2
bytes long. Parameter byte fields 2 through 9 are each 4 bytes long.
The supported fields are listed in Table 73.
WARNING
Table 73. Data Compression Log Parameters

Parameter Code
(in Hex)
0000h Read Compression ratio x 100 2
0001h Write Compression Ratio x 100 2
0002h Megabytes transferred to server 4
0003h Bytes transferred to server 4
0004h Megabytes read from tape 4
0005h Bytes read from tape 4
0006h Megabytes transferred from server 4
0007h Bytes transferred from server 4
0008h Megabytes written to tape 4
0009h Bytes written to tape 4
Parameters 2 through 9 occur as pairs that represent a large number of bytes

transferred. The first 4-byte parameter represents the number of whole megabytes
transferred, rounded to the nearest megabyte. The second 4-byte parameter
represents the difference between this number of megabytes and the actual number
of bytes. This may be a signed quantity.
This data may be reset, but may not be written.

Log Page 33h: Write Errors (Not Ultrium 1 or Ultrium 2
non-RoHS)
This page contains detailed counters related to write operations. This page is reset
when a cartridge is loaded.
Note: When multiple errors occur on a data set, the counter that is updated is
generally based on the first error detected.
WARNING
Table 74. Write Errors log parameter codes

Parameter Code
(in Hex)
Data sets Corrected: ECC is done by hardware. This is driven by an excessive
0000h CQs rewritten condition (TBD). Each count represents on data set in error that 2
was successfully corrected and written.
Servo Transients: ERP action was required because of a servo detected error and
the first retry was successfully in place (stop write without backhitch, that is,
0001h 2
servo write skip). Each count represents one data set in error that was
successfully recovered and written.
Data Transients: ERP action was required because of a readback check or ECC
detected error and the first retry was successfully in place (no backhitch). Each
0002h 2
count represents one data set in error that was successfully recovered and
written.
Velocity Events: A velocity control problem occurred. Each count represents on
0003h occurrence, not just the count of affected data sets. Counts may include 2
occurrences from both temporary and permanent errors.
Servo Acquisition Temps: A servo error (servo dropout or off-track shutdown)
was detected while trying to acquire a DSS or data set at the beginning of a
0004h write append sequence (motion). ERP action was required, and servo transient 2
condition criteria were not met. Each count represents one data set in error that
was successfully recovered and written.
Data Acquisition Temps: During read-back check, the read channel failed to
acquire a DSS or data set at the beginning of a write append sequence and no
0005h servo error was reported. ERP action was required, and read/ECC transient 2
condition criteria were not met. Each count represents one data set in error that
was successfully recovered and written.
Servo Temps: A servo error (servo dropout or off-track shutdown) was detected
while writing data. ERP action was required, and servo transient condition
0006h 2
criteria were not met. Each count represents one data set in error that was
successfully recovered and written.
Data Temps: An uncorrectable error, CRC error, instantaneous speed variation
(ISV) error, or no ending burst error occurred during readback check of a data
0007h set, and no servo error was reported. ERP action was required, and 2
readback/ECC transient condition criteria were not met. Each count represents
one data set in error that was successfully recovered and written.
Total Retries: The count of the total number of ERP actions. Each count
0008h represents one occurrence, not just one time per data set. Counts may include 2

Table 74. Write Errors log parameter codes (continued)
Parameter Code
(in Hex)
0009h Vendor-Reserved 2
000Ah Vendor-Reserved (Bellcord) 2
Servo Skip Events: The count of long servo write skips, extended DSS or long
spaces between data sets written. This is generally servo write skips, but may
000Bh also include other write scenarios. Each count represents one occurrence, not one 2
count per block. Counts may include occurrences from both temporary and
permanent errors.
Housekeeping Events: The count of write problems in the Housekeeping Data
000Ch set Region. Each count represents one occurrence, not just one time per data set. 2
Counts may include occurrences from both temporary and permanent errors.
FID Events: The count of write problems while processing the FID. Each count
000Dh represents one occurrence, not just one time per data set. Counts may include 2
000Eh Vendor-Reserved (Blocks Lifted) 2
Data set Underrun: The number of times that the drive overran the buffer
000Fh processing capability and had to stop and restart during a write. Each count 2
represents one occurrence, not just one time per write.
Servo Position Events: The number of servo detected positional compare
0011h 2
discrepancies.

Log Page 34h: Read Forward Errors (Not Ultrium 1 or Ultrium
2 non-RoHS)
This page contains detailed counters related to read operations. This page is reset
when a cartridge is loaded.
Note: When multiple errors occur on a data set, the counter that is updated is
generally based on the first error detected. ERP counters indicate which
specific ERP methods were successfully employed.
WARNING
Table 75. Read Forward Error Counters log parameter codes

Parameter Code
(in Hex)
Data sets Corrected: ECC is done by hardware. Each count represents one
0000h 2
data set in error that was successfully corrected and read.
Servo Transients: ERP action was required because of a servo detected error
0001h and the first retry was successfully in place. Each count represents one data 2
set in error that was successfully recovered and read.
Data Transients: ERP action was required because of a read channel or ECC
0002h detected error and the first retry was successfully in place. Each count 2
represents one data set in error that was successfully recovered and read.
Velocity Events: A velocity control problem occurred. Each count represents
0003h one occurrence, not just the count of affected data sets. Counts may include 2
occurrences from bot temporary and permanent errors.
Servo Acquisition Temps: A servo error (servo dropout or off track
shutdown) was detected while trying to acquire an initial DSS or data set.
0004h ERP action was required, and servo transient condition criteria were not 2
met. Each count represents one data set in error that was successfully
recovered and read.
Data Acquisition Temps: The read channel failed to acquire an initial DSS or
data set, and no servo error was reported. ERP action was required, and
0005h 2
read/ECC transient condition criteria were not met. Each count represents
one data set in error that was successfully recovered and read.
Servo Temps: A servo error (servo drop out) was detected while reading a
data set. ERP action was required, and servo transient condition criteria
0006h 2
were not met. Each count represents one data set in error that was
successfully recovered and read.
Data Temps: An uncorrectable error, CRC error, or no ending burst error
occurred while reading a data set, and no servo error was reported. ERP
0007h action was required, and read/ECC transient condition criteria were not 2
met. Each count represents one data set in error that was successfully
recovered and read.
Sequence Errors: A data set number out of sequence was encountered, and
no Servo or read/ECC error reported. ERP action was required, and no
0008h 2
transient condition criteria were not met. Each count represents one data set
in error that was successfully recovered and read.
0009h Vendor-Reserved (ERP Read Opposite) 2

Table 75. Read Forward Error Counters log parameter codes (continued)
Parameter Code
(in Hex)
000Ah Vendor-Reserved (ERP Tension Adjust Hi) 2
000Bh Vendor-Reserved (ERP Tension Adjust Lo) 2
ERP Servo Adjust Hi: The data set was recovered by reading with servo
000Ch off-track variations. Each count represents one data set in error that was 2
ERP Servo Adjust Lo: The data set was recovered by reading with servo
000Dh off-track variations. Each count represents one data set in error that was 2
000Eh Vendor-Reserved (ERP Dead Reckon Nominal) 2
000Fh Vendor-Reserved (ERP Dead Reckon Hi) 2
0010h Servo AGA Gain ERP (read only mode) 2
0011h Vendor-Reserved (ERP Filter Coefficients) 2
0012h Servo Opposite Gap ERP (read only mode) 2
0013h Vendor-Reserved (ERP Dataflow Clock Adjust) 2
Total Retries: The count of the total number of ERP actions. Each count
0015h represents one occurrence, not just one time per data set. Counts may 2
include occurrences from both temporary and permanent errors.
0016h Match Filter ERP (read mode use) 2
Housekeeping Events: The count of read problems in the Housekeeping
data set Region. Each count represents one occurrence, not just one time per
0017h 2
data set. Counts may include occurrences from both temporary and
permanent errors.
0018h Vendor-Reserved (Cartridge Init Errors) 2
data set Overrun: The number of times that the drive overran the buffer
0019h processing capability and had to stop and restart during a read. Each count 2
represents one occurrence, not just one time per read.
001Ah Vendor-Reserved 2
Servo Skip Events: The count of extended DSS or long spaces between data
sets read. This may include servo write skips, but may also include other
001Bh write scenarios. Each count represents one occurrence, not one count per 2
block. Counts may include occurrences from both temporary and permanent
errors.
001Ch Vendor-Reserved 2
FID Events: The count of write problems while processing the FID. Each
001Dh count represents one occurrence, not just one time per data set. Counts may 2
include occurrences from both temporary and permanent errors.
Servo Position Events: The number of servo detected positional compare
001Eh 2
discrepancies.

Log Page 37h: Performance Characteristics (Ultrium 4)
This page includes various performance and capacity measurements across the
operation of the drive. Some fields are normalized qualitative measures while
others are quantitative. This page uses the Subpage Code mechanism (see “LOG
SENSE” on page 50) to select which groups of counters to return. This page has
three scopes controlled by the Subpage Code field. Each has different reset
characteristics which are described under the Subpage Code field description
below.
The subpage field in “LOG SENSE” on page 50 is used as follows:

Bit Description
7-6 Scope
Value Description
0h Transient values: reset on Log Select [all Subpages are reset]
1h Mount values: reset on load
2h Lifetime values: reset on device power on or device reset (not
target reset)
3h Vendor-Reserved
5-4 Level
Value Description
0h Return summary counters
1h Return basic counters
2h Return advanced counters
3-0 Group
Value Description
0h All groups
1h Host Interface
2h Buffer
3h Medium
4h Capacity
5h Load/Unload
6h Servo
The Subpage field in “LOG SELECT” on page 49 may be set to 00h when the page
code field is 37h. This operation will reset all group and local counters in the
transient scope. Other scope cannot be explicitly reset.
The individual log subpage and parameter codes are described in the following
table. Note that the counters which are returned depends on the Level and Group
fields in the subpage. A group value of 0h will return all counters of a level less
than or equal to that specified.
In the following tables, multiple counter codes may be represented by a single row.
There will be an aspect symbol in the counter code such as ’p’, ’q’, ’s’, or ’?’. The
Aspect(s) column indicates which of the following values applies to the given
code(s).
Aspect Definition
p=0 primary interface all ports (totals)
p=1 primary interface port 0
p=2 primary interface port 1

p=A automation interface (RS-422)
q=1 Non-Ready: NOTE: These commands include ALL commands which are
processed when the drive is in a Not Ready state.
q=2 Head-of-Queue: NOTE: These commands are commands which may be
processed in any order. Such commands include: Inquiry, Report LUNs,
Test Unit Ready and Request Sense. These counts are updated only when
the drive is in a Ready state.
q=2 Head-of-Queue: NOTE: These commands are commands which may be
processed in any order. Such commands include: Inquiry, Report LUNs,
Test Unit Ready and Request Sense. These counts are updated only when
q=3 Read: NOTE: This aspect has more features detailed below.
q=4 Write: NOTE: This aspect has more features detailed below.
q=5 Sync: NOTE: This aspect has more features detailed below.
q=6 Seek: NOTE: This aspect has more features detailed below.
q=7 Non-Medium: NOTE: These commands are command issued to LUN 0
which are not in any other applicable category. These include many
commands such as Log Sense, Log Select, Read Buffer, Reserve, etc. These
counts are updated only when the drive is in a Ready state.
q=8 Non-LUN0, Non-Ready: NOTE: These commands include ALL commands
which are processed by a LUN other than LUN 0 when the drive is in a
Not Ready state.
q=9 Non-LUN0, Head-of-Queue: NOTE: These commands are commands
which may be processed in any order. Such commands include: Inquiry,
Report LUNs, Test Unit Ready and Request Sense. These counts are
updated only when the drive is in a Ready state.
q=A Non-LUN0: NOTE: These commands include any commands processed by
a LUN other than LUN 0. These counts are updated only when the drive is
in a Ready state.
s=1 Speed 1: Highest read/write speed
s=2 Speed 2: Second highest read/write speed
s=3 Speed 3: Third highest read/write speed
s=4 Speed 4: Fourth highest read/write speed
s=5 Speed 5: Fifth highest read/write speed
s=6 Speed 6: Slowest read/write speed
s=F High speed locate (not read/write capable)
?=1 write - write phase without host holdoff. NOTE: if no (paused) data is
supported for a particular counter, this aspect will include all write
information
?=2 write (paused) - write phase while the host is being held off (buffer full)
?=3 read - read phase without host holdoff. NOTE: if no (paused) data is
supported for a particular counter, this aspect will include all read
information
?=4 read (paused) - read phase while the host is being held off (buffer empty)

?=5 position - during the processing of a seek operation
?=6 load - during the processing of an load operation
?=7 unload - during the processing of an unload operation
?=F other - not in an above phase
Table 76. Log Page 37h: Performance Characteristics: Quality Summary
Code Aspect(s) Name : Description Unit Size Level Group
Drive Efficiency: Overall measure of the
0000h drive’s condition. 00h is unknown, from 01h rating 1 0 0
(best) to FFh (worst)
Media Efficiency: Overall measure of the
0001 currently mounted media’s condition. 00h’ is rating 1 0 0
unknown, from 01h (best) to FFh (worst)
Primary Interface Efficiency: Overall measure
0010h of the interface (to the host) condition. 00h is rating 1 0 0
unknown, from 01h (best) to FFh(worst)
Primary Interface Port 0 Efficiency: Overall
measure of the per port interface (to the host)
0011h rating 1 0 0
condition. 00h is unknown, from 01h (best) to
FFh (worst)
Primary Interface Port 1 Efficiency: Overall
measure of the per port interface (to the host)
0012h rating 1 0 0
condition. 00h is unknown, from 01h (best) to
FFh (worst)
Library Interface Efficiency: Overall measure of
001Ah the interface (to the library) condition. 00h is rating 1 0 0
unknown, from 01h (best) to FFh (worst)
Table 77. Log Page 37h: Performance Characteristics: Device Usage

0040h Time: Amount of entire sample duration msec 6 2 5
Medium Empty Time: Duration without a tape
0041h msec 6 2 5
present
Medium Insert Time: Duration from cartridge
0042h msec 6 2 5
insert to load
Medium Mount Time: Total time from start of
0043h msec 6 2 5
cartridge load until cartridge ejected
Medium Load Time: Total time from start of
0044h msec 6 2 5
cartridge load to load complete (ready)
Medium Ready Time: Total time from load
0045h msec 6 2 5
complete (ready) to start of unload
Medium Eject Time: Time from start of unload to
0046h msec 6 2 5
unload complete
Medium Extract Time: Time from cartridge
0047h msec 6 2 5
unloaded to removed
Medium Dwell Time: Time from cartridge
unloaded to (re)loaded.
Note: This may include time which cannot be
0048h msec 6 2 5
determined as dwell or extract (when time is
queried with a cartridge remains in the unloaded
position)

Table 77. Log Page 37h: Performance Characteristics: Device Usage (continued)
Medium Clean Time: Time from cleaner recognized
0049h msec 6 2 5
to eject complete
Medium Empty Count: Number of times tape was
0051h count 4 2 5
fully removed
Medium Insert Count: Number of cartridge
0052h count 4 2 5
insertions to load position detected
Medium Mount Count: Number of mount
0053h count 4 2 5
operations
0054h Medium Load Count: Number of load operations count 4 2 5
Medium Ready Count: Number of ready
0055h count 4 2 5
transitions
0056h Medium Eject Count: Number of unloads count 4 2 5
Medium Extract Count: Number of times tape was
0057h count 4 2 5
extracted
Medium Dwell Count: Number of times tape was
0058h count 4 2 5
reloaded (from unload)
Medium Clean Count: Number of recognized
0059h cleaner loads (does not indicate successful cleans, count 4 2 5
tape may be expired)
Table 78. Log Page 37h: Performance Characteristics: Host Commands

0q00h 12(3456)789A ... Count count 4 2 1
0q01h 12(3456)789A ... Timing msec 6 2 1
0q02h 12(3456)789A ... Relative Time % * 65536 4 1 1
0q04h 12(3456)789A ... Transfer Count to Host (in) count 4 2 1
0q05h 12(3456)789A ... Transfer Byte Count to Host (in) bytes 8 2 1
0q06h 12(3456)789A ... Transfer Timing to Host (in) msec 6 2 1
0q08h 12(4)789A ... Transfer Count from Host (out) count 4 2 1
0q09h 12(4)789A ... Transfer Byte Count from Host (out) bytes 8 2 1
0q0Ah 12(4)789A ... Transfer Timing from Host (out) msec 6 2 1
Note: Read type host commands include Read, Verify and Read Reverse (not all of these may be supported). These
counts are updated only when the drive is in a Ready state
Read Count: Number of blocks processed to the
0300h blocks 6 2 1
host by read type commands
Read Timing: Amount of time processing read type
commands.
Note: Due to device specific performance path
0301h resources, this may not reflect the actual time msec 6 2 1
spend processing commands, but may reflect the
amount of time where read commands could be
processed.
Read Relative Time: Ratio of time spent reading
0302h % * 65536 4 1 1
with respect to Medium Ready Time
0304h ... Transfer Count to Host (in) count 6 2 1
0305h ... Transfer Byte Count to Host (in) bytes 8 2 1

Table 78. Log Page 37h: Performance Characteristics: Host Commands (continued)
0306h ... Transfer Timing to Host (in) msec 6 2 1
Read Performance Efficiency: Ratio of performance
03D0h read type commands with respect to all read type % * 65536 4 1 1
commands
Read Filemark (perf) Relative Time: Amount of
03D4h time spent sending filemark encountered status to % * 65536 4 2 1
the host with respect to time Read Timing.
Note: Write type host commands include Write and Write Filemarks [not including any synchronization portion].
These counts are updated only when the drive is in a Ready state.
Write Count: Number of blocks processed from the
0400h blocks 6 2 1
host by write type commands
Write Timing: Amount of time processing write
type commands.
Note: Due to device specific performance path
0401h resources, this may not reflect the actual time msec 6 2 1
spend processing commands, but may reflect the
amount of time where write commands could be
processed.
Write Relative Time: Ratio of time spent writing
0402h % * 65536 4 1 1
with respect to Medium Ready Time
0408h ... Transfer Count from Host (out) count 6 2 1
0409h ... Transfer Byte Count from Host (out) bytes 8 2 1
040Ah ... Transfer Timing from Host (out) msec 6 2 1
Write Performance Efficiency: Ratio of performance
04D0h write commands with respect to all write type % * 65536 4 1 1
commands
Write Filemark Relative Time: Amount of time
04D4h % * 65536 4 2 1
spent writing filemarks.
Note: Sync type host commands include Write Filemarks [non-immediate]. Implicit sync type commands include
mode changes while writing, non-buffered mode and idle time based syncs. These counts are updated only when
Sync Count [Host]: Number of host sync
0500h operations (non-immediate Write Filemarks, count 4 2 1
non-buffered writes)
Sync Timing [Host]: Amount of time processing
0501h msec 6 2 1
host sync commands
Sync Relative Time [Host]: Ratio of time spent
0502h processing host sync commands with respect to % * 65536 4 1 1
Medium Ready Time
Sync Count [Implicit]: Number of implicit sync
05D1h commands (time based flushes, mode change count 4 2 1
flushes)

Sync Timing [Implicit]: Amount of time processing
05D2h msec 6 2 1
implicit sync commands
Sync Relative Time [Implicit]: Ratio of time spent
05D4h processing implicit sync commands with respect to % * 65536 4 1 1
Medium Ready Time
Note: Seek type host commands include Space, Locate, and Rewind. These counts are updated only when the drive
is in a Ready state.
Seek Count: Number of positioning host
0600h count 4 2 1
commands
Seek Timing: Amount of time spent processing
0601h msec 6 2 1
host positioning commands
Seek Relative Time: Ratio of time spend processing
0602h host seek commands with respect to Medium % * 65536 4 1 1
Ready Time
Seek Block Count: Number of blocks processed in
06D1h blocks 4 2 1
host positioning commands
0Cp0h 012A Command Count count 6 2 1
0Cp1h 012A Command Timing msec 6 2 1
0Cp2h 012A Command Relative Time % * 65536 4 2 1
0Cp4h 012A Command Transfer Count to Host (in) count 6 2 1
0Cp5h 012A Command Transfer Byte Count to Host (in) bytes 8 2 1
0Cp6h 012A Command Transfer Timing to Host (in) msec 6 2 1
0Cp8h 012A Command Transfer Count from Host (out) count 6 2 1
0Cp9h 012A Command Transfer Byte Count from Host (out) bytes 8 2 1
0CpAh 012A Command Transfer Timing from Host (out) msec 6 2 1
0CpCh 012A Command Queue Count count 6 2 1
0CpDh 012A Command Queue Latency msec 6 2 1
0CpEh 012A Command Queue Relative Time % * 65536 4 2 1
0Dp1h 12A Port Throughput Rate Maximum Bursting bytes/sec 6 2 1
0Dp2h 12A Port Throughput Rate Maximum Sustained bytes/sec 6 2 1
0Dp3h 12A Port Throughput Rate bytes/sec 6 2 1
0Dp4h 12A Port Throughput Efficiency % * 65536 4 1 1
0Dp7h 12A Port Rate Changes count 4 2 1
Average Command Latency: Average amount of
time commands were queued waiting for execution
0DF0h usec 4 1 1
with respect to all commands (including unqueued
commands which executed immediately).
Average Dequeue Latency: Average amount of
time commands were queued waiting for execution
0DF1h usec 4 1 1
with respect to commands which where queued
(not executed immediately).

0DF8h Long Queue Latency Count [>1 sec] count 4 2 1
0DF9h Long Queue Latency Count [>10 sec] count 4 2 1
0DFAh Long Queue Latency Count [>100 sec] count 4 2 1
0DFBh Long Queue Latency Count [>1000 sec] count 4 2 1
Table 79. Log Page 37h: Performance Characteristics: Host Initiators

Note: In dual ported configurations it is possible for the same host to be identified as a different initiator when
using a different logical or physical path. This can occur in failover or load balancing applications.
Active Initiator Count: Number of initiators which
0E00h count 4 2 1
executed one or more commands.
Primary Initiator: Ratio of commands issued by the
0E01h initiator which is issuing the most commands with % * 65536 4 2 1
respect to all initiators.
Secondary Initiator: Ratio of commands issued by
0E02h the initiator which is issuing the second most % * 65536 4 2 1
commands with respect to all initiators.
Current Initiator: Ratio of commands issued by
0E03h this (the querying) initiator with respect to all % * 65536 4 2 1
initiators.
Table 80. Log Page 37h: Performance Characteristics: Host Recovery (by port)
0Fp0h 12A Transfer Recoveries [by port] count 4 2 1
0Fp1h 12A Transfer Recover Time [by port] msec 6 2 1
0Fp2h 12A Resource Recoveries [by port] count 4 2 1
0Fp3h 12A Reset Count [by port] count 4 2 1
0Fp8h 12A Abort Count [by port] count 4 2 1
0Fp9h 12A Abort Time [by port] msec 6 2 1
Table 81. Log Page 37h: Performance Characteristics: Mode Phase Timing Windows
1000h Write Cycles count 4 2 2
1001h Write Pauses count 4 2 2
1010h Write Cycle Time msec 6 2 2
Write Cycle Relative Time: Ratio of time in write
1020h % * 65536 4 1 2
mode with respect to Medium Ready Time.
1021h Write Setup Relative Time % * 65536 4 2 2
1022h Write Ready Relative Time % * 65536 4 1 2
1023h Write Pause Relative Time % * 65536 4 1 2
1024h Write Exit Relative Time % * 65536 4 2 2
200h Read Cycles count 4 2 2
1201h Read Pauses count 4 2 2

Table 81. Log Page 37h: Performance Characteristics: Mode Phase Timing Windows (continued)
1210h Read Cycle Time msec 6 2 2
Read Cycle Relative Time: Ratio of time in write
1220h % * 65536 4 1 2
mode with respect to Medium Ready Time.
1221h Read Setup Relative Time % * 65536 4 2 2
1222h Read Ready Relative Time % * 65536 4 1 2
1223h Read Pause Relative Time % * 65536 4 1 2
1224h Read Exit Relative Time % * 65536 4 2 2
1225h Read Traverse EM Relative Time % * 65536 4 2 2
1400h Position Count count 4 2 2
1410h Position Time msec 6 2 2
Position Relative Time: Ratio of time spent
1420h physically and logically positioning with respect to % * 65536 4 1 2
Medium Ready Time.
1430h Position Relative Rate bytes/sec 4 1 2
1480h Position Count (Media) count 4 2 2
1490h Position Time (Media) msec 6 2 2
Position Relative Time (Media): Ratio of time spent
14A0h physically positioning media with respect to % * 65536 4 1 2
Medium Ready Time.
14B0h Position Relative Rate (Media) bytes/sec 4 1 2
Position Buffer Hits: Ratio of positioning
14F0h operations where targets were already present in % * 65536 4 1 2
the buffer.
Flush Count: Number of low level buffer write
flush operations. These may include operations
1500h count 4 1 2
which only affect the buffer and do not involve
media motion.
Flush Time: Time spend executing operations
1510h msec 6 2 2
counted by Flush Count.
Flush Relative Time: Ratio of time spent flushing
1520h % * 65536 4 1 2
with respect to Medium Ready Time.
Flush Count (Media): Number of low level buffer
1580h write flush operations which involve [or continue] count 4 1 2
media motion.
Flush Time (Media): Time spent executing
1590h msec 6 2 2
operations counted by Flush Count (Media).
Flush Relative Time (Media): Ratio of time spent
15A0h flushing to media with respect to Write Cycle % * 65536 4 1 2
Time.
Flush Buffer Hits: Ratio of flush operations which
15F0h % * 65536 4 1 2
required media motion.
2000h Media Idle msec 6 2 3
2001h Media Write msec 6 2 3
2002h Media Read msec 6 2 3
2003h Media Erase msec 6 2 3
2004h Media Position msec 6 2 3

20?0h 1234567F Media Phase Timing msec 6 2 3
20?1h 1234567F Media Phase Cycles count 4 2 3
Wrap Change Count: Total number of wrap
2?10h 1234567F count 4 2 3
changes.
Band Change Count: Total number of changes to
2?11h 1234567F count 4 2 3
different servo bands.
2?50h 13 Datarate Performance Impacting ERPs % * 65536 4 1 3
2?51h 13567F Performance Impacting ERPs % * 65536 4 1 3
2?52h 1234567F Performance Impact by ERPs % * 65536 4 1 3
2?60h 135 Uncompressed Data bytes 8 2 2
2?61h 135 Compressed Data bytes 8 2 2
2?62h 135 Padded Data bytes 8 2 2
2?63h 135 Degate Data bytes 8 2 2
2?68h 135 Data sets Processed data sets 4 2 2
2?6Ch 13 Compression Ratio % * 65536 4 1 2
2?71h 13F Compressed Data (Medium) bytes 8 2 3
2?72h 13F Padded Data (Medium) bytes 8 2 3
2?80h 13 Maximum Host Transfer Rate bytes/sec 4 1 2
2?81h 13 Average Host Transfer Rate bytes/sec 4 1 2
2?82h 13 Average Host Buffer Rate bytes/sec 4 1 2
2?83h 13 Window Host Buffer Rate bytes/sec 4 1 2
2?84h 13 Host Buffer Efficiency % * 65536 4 1 2
2?85h 13 Window Buffer Efficiency % * 65536 4 1 2
2?88h 13 Average Host Transfer Length bytes 4 2 2
2?8Ch 1 Average Host Sync Length bytes 6 2 2
2?90h 13 Maximum Comp Transfer Rate bytes/sec 4 1 2
2?91h 13 Average Comp Transfer Rate bytes/sec 4 1 2
’2?92h 13 Average Comp Buffer Rate bytes/sec 4 1 2
2?93h 13 Window Comp Buffer Rate bytes/sec 4 1 2
2?94h 13 Comp Buffer Efficiency % * 65536 4 1 2
2?95h 13 Window Comp Buffer Efficiency % * 65536 4 1 2
2?98h 13 Average Comp Transfer Length bytes 4 2 2
2?9Ch 1 Average Comp Sync Length bytes 6 2 2
2?A0h 13 Maximum Tape Transfer Rate bytes/sec 4 1 2
2?A1h 13 Average Tape Buffer Rate bytes/sec 4 2 2
2?A2h 13 Window Tape Buffer Rate bytes/sec 4 2 2
2?A3h 13 Moving Tape Buffer Rate bytes/sec 4 2 2
2?A4h 13 Window Tape Buffer Efficiency % * 65536 4 2 2
2?A5h 13 Moving Tape Buffer Efficiency % * 65536 4 2 2

Tape Buffer Efficiency: Ratio of amount of time we
are usefully moving and ready with respect to
amount of time the buffer is able to process data.
2?A6h 13 A ratio larger than 1 indicates the compressed host % * 65536 4 2 2
data is arriving faster than the native device rate.
Lower values indicate the device has underutilized
host bandwidth.
Tape Thrashing: Ratio of amount of time we are
2?A7h 13F accelerating, decelerating or backhitching with % * 65536 4 2 2
respect to the time in mode.
Tape Efficiency: Ratio of amount of time we are
2?A8h 13F usefully moving and ready with respect to the % * 65536 4 1 2
time in mode.
2?F0h 13 Speed Changes count 4 2 2
2?F1h 13 Speed Forced count 4 2 2
Table 82. Log Page 37h: Performance Characteristics: Servo Speed Characteristics
5Fs0h 123456F Servo Speed Relative Time % * 65536 4 2 6
Table 83. Log Page 37h: Performance Characteristics: Static Capacity

7000h Static Capacity Efficiency % * 65536 4 1 4
7010h Static Data sets Media data sets 4 2 4
7011h Static Data sets Used data sets 4 2 4
7020h Static Distance Media mm 8 2 4
7021h Static Distance Used mm 8 2 4
Table 84. Log Page 37h: Performance Characteristics: Active Capacity

7?00h 13F Active Capacity Efficiency % * 65536 4 1 4
7?10h 13F Active Sync Loss % * 65536 4 2 4
7?11h 13F Active Skip Loss % * 65536 4 2 4
7?12h 13F Active DSS Loss % * 65536 4 2 4
7?13h 13F Active CQs Loss (on-the-fly) % * 65536 4 2 4
7?21h 13F Active Distance Skip mm 8 2 4
7?22h 13F Active Distance DSS mm 8 2 4
7?23h 13F Active Distance CQs (on-the-fly) mm 8 2 4
7?2Fh 13F Active Distance Total mm 8 2 4
Note: The parameters in this page should be dynamically parsed as some

parameters may not be present and other parameters may be inserted.

Log Page 38h: Blocks/Bytes Transferred (Not Ultrium 1 or
Ultrium 2 non-RoHS)
This page is reset when a cartridge is loaded.
WARNING
Table 85. Blocks/Bytes Transferred log parameter codes

Parameter Code
(in Hex)
Host Write Blocks Processed: Each count represents a block processed across the
0000h 4
host interface during a Write. The count does not include ERP retries.
Host Write Kilobytes Processed: Each count represents a kilobyte (1024 bytes)
processed across the host interface during a Write. The count does not include
0001h 4
ERP retries. This count may be divided by Device Write Kilobytes Processed,
0005h, to calculate an approximate write comparison ratio.
Host Read Blocks Processed: Each count represents a block processed across the
0002h 4
host interface during a Read. The count does not include ERP retries.
Host Read Kilobytes Processed: Each count represents a kilobyte (1024 bytes)
processed across the host interface during a Read. The count does not include
0003h 4
ERP retries. This count may be divided by Device Read Kilobytes Processed,
0007h, to calculate an approximate read compression ratio.
Device Write Data sets Processed: Each count represents a data set processed on
0004h 4
the medium. The count does not include ERP retries.
Device Write Kilobytes Processed: Each count represents a kilobyte (1024 bytes)
0005h processed on the medium. The count does not include ERP retries or any tape 4
formatting overhead bytes.
Device Read Data sets Processed: Each count represents a data set processed
0006h 4
from the medium. The count does not include ERP retries.
Device Read Kilobytes Processed: Each count represents a kilobyte (1024 bytes)
0007h processed from the medium. The count does not include ERP retries or any tape 4
Device Write Data sets Transferred: Each count represents a data set processed
0008h 4
on the medium. The count includes ERP retries.
Device Write Kilobytes Transferred: Each count represents a kilobyte (1024 bytes)
0009h processed on the medium. The count includes ERP retries and any tape 4
Device Read Data sets Transferred: Each count represents a data set processed
000Ah 4
from the medium. The count includes ERP retries.
Device Read Kilobytes Transferred: Each count represents a kilobyte (1024 bytes)
000Bh processed from the medium. The count includes ERP retries and any tape 4
Nominal Capacity of Partition: The nominal capacity of the current partition (in
000Ch 4
kilobytes).
Fraction of Partition Traversed: The fractional part of the current partition
000Dh 1
traversed (N/255).

Table 85. Blocks/Bytes Transferred log parameter codes (continued)
Parameter Code
(in Hex)
Nominal Capacity of Volume: The nominal capacity of the mounted volume in
000Eh 4
kilobytes. This is not sensitive to current position.
Fraction of Volume Traversed: The fractional part of the mounted volume
000Fh 1
traversed (N/255).
Remaining Capacity of Volume: The nominal unwritten remaining capacity of
0010h 4
the mounted volume (in kilobytes). This is not sensitive to current position.
Remaining Capacity of Partition: The nominal unwritten remaining capacity of
0011h 4
the current partition (in kilobytes). This is not sensitive to current position.
Note: When the tape does not have a valid EOD, or if a tape is not loaded, the drive will attempt to return a value
of ’all ones’ (-1).

Log Page 39h: Host Port 0 Interface Errors (Not Ultrium 1 or
Ultrium 2 non-RoHS)
This page shows the count of errors occurring on primary port 1 (while the device
is active on the interface).
WARNING
Table 86. Host Port Interface log parameter codes

Parameter Code
(in Hex)
Host Protocol Errors: Each count represents one occurrence. Counts may include
occurrences from both temporary and permanent errors. (On SAS drives, this
0000h 2
parameter is not returned. See Table 68 on page 74, Protocol-Specific log page for
SAS protocol error counts).
Host Aborts: Each count represents one occurrence. Counts may include
0007h 2
Host Resets: Each count represents one occurrence. Counts may include
0008h 2
000Ah Vendor-Reserved 2
Host Recoveries: Each count represents one occurrence. Counts may include
occurrences from both temporary and permanent errors. An example of a host
0010h 4
recoveries is a Sequence Retransmission Request (SRR) on Fibre Channel drives.
(Not returned on SAS drives)

+ Log Page 3Ah: Drive Control Verification (Not in Ultrium 1)
+ This page is for special drive control mode verification. This page is not included
+ in the list of supported pages (page code 00h) and is not intended for general use.

Log Page 3Bh: Host Port 1 Interface Errors (Not Ultrium 1 or
Ultrium 2 non-RoHS)
This page shows the count of errors occurring on primary port 2 (while the device
is active on the interface).
Note: The parameters are identical to those found in “Log Page 39h: Host Port 0
Interface Errors (Not Ultrium 1 or Ultrium 2 non-RoHS)” on page 97, except
this data is record for incidents that occur on primary port 2.
WARNING

Log Page 3Ch: Drive Usage Information
Log Page 3Ch only has valid data returned when the drive is configured for use in
an eServer system (i.e., The Product ID returned in Standard Inquiry is "HH LTO
Gen x" where "x" is a number).

Log Page 3Dh: Subsystem Statistics (Not Ultrium 1 or Ultrium
2 non-RoHS)
The following counters all deal with subsystem statistics and errors. Most of the
counters on this page are never reset. Most counters are maintained in VPD and
persist across Log Selects, Log Sense, Power On Resets, and even microcode
download. Lifetime values are written to VPD every eight operating hours when
the drive is in a not ready state. The counters lock at maximum values.
WARNING
Table 87. Subsystem Statistics log parameter codes

Parameter Code
(in Hex)
Volume Lifetime Mounts: The total number of successful cartridge unloads
0020h performed during the lifetime of a cartridge. This field may not be updated for 4
mounts that occur with the volume physically write protected.
Volume Lifetime Megabytes Written: The total amount of data in Megabytes
written during the lifetime of the cartridge. On each unload, an approximate
value is calculated and stored by rounding up to the nearest Megabyte. These
0021h 4
bytes are counted as they are processed to the medium (compressed bytes), not
at the host interface. This field may not be updated during mounts that occur
with the volume physically write protected.
Volume Lifetime Megabytes Read: The total amount of data in Megabytes read
during the lifetime of the cartridge. On each unload, an approximate value is
calculated and stored by rounding up to the nearest Megabyte. These bytes are
0022h 4
counted as they are processed from the medium (compressed bytes), not at the
host interface. This field may not be updated during mounts that occur with the
volume physically write protected.
Drive Lifetime Mounts: The total number of successful cartridge unloads
0040h 4
performed during the lifetime of the drive.
Drive Lifetime Megabytes Written: The total amount of data in Megabytes
written during the lifetime of the drive. On each unload, an approximate value is
0041h calculated and stored by rounding up to the nearest Megabyte. These bytes are 4
counted as they are processed to the medium (compressed bytes), not at the host
interface.
Drive Lifetime Megabytes Read: The total amount of data in Megabytes read
during the lifetime of the drive. On each unload, an approximate value is
host interface.
Clean Lifetime Mounts: The total number of successful cleaner cartridge
0060h 4
operations performed during the lifetime of the drive.
Megabytes Written Since Clean: The total amount of data in Megabytes written
since the last successful clean operation. On each unload, an approximate value
0061h is calculated and stored by rounding up to the nearest Megabyte. These bytes are 4
host interface.

Table 87. Subsystem Statistics log parameter codes (continued)
Parameter Code
(in Hex)
Megabytes Read Since Clean: The total amount of data in Megabytes read since
the last successful clean operation. On each unload, an approximate value is
host interface.
Mounts Since Clean: The total number of mounts performed since the last
0063h 4
successful clean operation.
Library Interface Messages Received: This counter is not stored in VPD and
0080h reflects messages received for which the drive sent a positive acknowledgement 4
since reset.
Library Interface Messages Transmitted: This counter is not stored in VPD and
0081h reflects messages since reset. This is incremented every time a message is sent 4
and does not imply that an acknowledgement was received.
Library Interface Resets: Count of hardware reset or logical re-initializations
0082h 4
during normal operation. This is a link reset and not a power-on reset.
Library Interface Buffer Errors: This includes buffer overrun or underrun
0083h 4
conditions. This includes both H/W and S/W buffers.
Library Interface Sync Errors: This is incremented any time an unexpected
0084h 4
character or set of characters occurs outside of a frame.
Library Interface Framing Errors: This is incremented any time there is a length
0085h 4
problem, a bad checksum, or a missing End character.
Library Interface Protocol Errors: This is incremented each time an extra or
0086h unexpected acknowledgement is received, as well as every time there are 4
consecutive messages with the same Message ID.
Library Interface Logical Errors: This is incremented every time a logic error is
0087h 4
seen in the received message (for example, Message Type error).
Library Interface Loader Failures: This counter reflects load/unload attempts
when the drive is in an incorrect state, or was otherwise unable to attempt
0088h 4
requested loader action. This is not a failure of a load/unload that was actually
attempted by the drive.
Library Interface NAKs Received: This is incremented every time a Negative
0089h 4
Acknowledgement is received.
Library Interface Acknowledgement Timeout: This is incremented every time a
008Ah timeout occurs while waiting for an acknowledgement (for example, Link 4
timeout).
Library Interface Application Layer Timeout: This is incremented every time the
008Bh 4
drive times out waiting for an application layer Response Message.
Drive Lifetime Write Perms: Total number of write permanent errors that
0090h 4
occurred on this drive.
Drive Lifetime Read Perms: Total number of read permanent errors that occurred
0091h 4
on this drive.
Drive Lifetime Load Perms: Total number of load permanent errors that occurred
0092h 4
on this drive.
Drive Lifetime Unload Perms: Total number of unload permanent errors that
0093h 4

Table 87. Subsystem Statistics log parameter codes (continued)
Parameter Code
(in Hex)
Drive Lifetime Write Temps: Total number of write temporary errors that
00A0h 4
Note: This is not a count of ERP actions taken, and certain transient errors may
not be included in this count.
Drive Lifetime Read Temps: Total number of read temporary errors that occurred
on this drive.
00A1h 4
Drive Lifetime Load Temps: Total number of load temporary errors that occurred
on this drive.
00A2h 4
Drive Lifetime Unload Temps: Total number of unload temporary errors that
00A3h 4
Lifetime Power On Seconds: Cumulative number of seconds that the drive has
been powered on.
0100h Note: Since this time is only periodically updated in non-volatile storage, it is 4
possible that this time may not be entirely accurate to the full resolution of the
counter.
Power On Seconds: Number of seconds since the drive was powered on or has
0101h 4
undergone a hard reset condition.
Reset Seconds: Number of seconds since the drive has undergone a soft reset
0102h 4
condition.

+ Log Page 3Eh: Engineering Use (Not Ultrium 1 or Ultrium 2
+ non-RoHS)
+ This page is for engineering use only and is not included in the list of supported
+ pages (page code 00h). As such, the counters on this page are not intended for
+ general use.

Log Page 3Eh[3Ch]: Drive Control Statistics (Not Ultrium 1 or
Ultrium 2 non-RoHS)
This page is for special drive control mode statistics. This page is not included in
the list of supported pages (page code 00h), and is not intended for general use.

MODE SELECT
The MODE SELECT command causes configuration data to be sent to the drive.
For the format of Mode data and supported Mode pages see “MODE SENSE” on
page 107. Both the 10-byte and 6-byte versions of the MODE SELECT command
are supported.
Table 88. 6-Byte MODE SELECT Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete PF Reserved SP
2 Reserved
3 Reserved
4 Parameter List Length
5 Control
Table 89. 10-Byte MODE SELECT Command

Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete PF Reserved SP
2 Reserved
3 Reserved
4 Reserved
5 Reserved
6 Reserved
7
8
9 Control
The Page Format (PF) field may be any value. However, the drive assumes that the
format is SCSI-2. The Save Pages (SP) field is only allowed to be set to 1 when
explicitly mentioned in the description of the specific mode page. The Parameter
List Length field should be set to the amount of data to be sent to the drive.
The Parameter List Length field should be set to the amount of data to be sent to
the drive. The Parameter List Length value is checked to ensure that it is less than
or equal to the sum of the lengths of all the supported mode pages. Any command
with a larger value results in a Check Condition status. The associated sense data
returns with a Sense Key of Illegal Request and an ASC/ASCQ of Invalid Field in
CDB (2400).
If any of the fields in the Mode pages are invalid, no parameters are altered, Check
Condition status is returned, the Sense Key is set to Illegal Request, and the
ASC/ASCQ is set to Invalid Field in Parameter List (2600).

MODE SENSE
The MODE SENSE command requests that the drive send its configuration data to
the initiator. Pages 01h, 02h, 0Fh, 10h, and 1Ch are supported. Mode Pages 18h
and 19h are supported only on Fibre Channel drives . A page 3Fh request returns
all supported pages.
Both the 10-byte and the 6-byte versions of the command are supported.
Table 90. 6-Byte MODE SENSE Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (1Ah)
1 Obsolete Reserved DBD Reserved
2 PC Page Code
3 Reserved
4 Allocation Length
5 Control
Table 91. 10-Byte MODE SENSE Command

Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved DBD Reserved
2 PC Page Code
3 Reserved
4 Reserved
5 Reserved
6 Reserved
7
: Allocation Length
8
9 Control
If the Disable Block Descriptors (DBD) field is set to 1, then no block descriptors
are returned with the mode data. If it is set to 0, then block descriptors are
returned.
The type of data returned is determined by the value of the Page Control (PC)
field as follows:
v 00b means the current configuration.
v 01b means the changeable bitmap (changeable = 1; unchangeable = 0).
v 10b means the default (power-on) values.
v 11b means the saved values. Because the drive does not support saved values,
the default values are returned.

The PC field only affects the mode parameters within the modes pages. It does not
affect the mode parameter header or the mode block descriptor. Within the mode
parameters, the PC field does not affect the PS field, the Page Code, or the
Additional Page Length fields. These fields will return the current values, as there
is no meaning if they are changed.
The Page Code must be set to the page to be returned.

Mode Data Format
Mode data returned by a MODE SENSE command or sent with a MODE SELECT
command consists of a Mode Parameter Header, an optional Mode Block
Descriptor, and zero or more Mode Parameter Pages.
Table 92. Mode Data Format
Bit
Byte 7 6 5 4 3 2 1 0
0
: Mode Parameter Header
m
m+1
: Mode Block Descriptor
m+8
m+9
: Mode Parameter Pages

Mode Parameter Header
Table 93. Mode Parameter Header 6-Byte Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Mode Data Length
1 Medium Type
2 WP Buffered Mode Speed
3 Block Descriptor Length
Table 94. Mode Parameter Header 10-Byte Command

Bit
Byte 7 6 5 4 3 2 1 0
0
: Mode Data Length
1
2 Medium Type
3 WP Buffered Mode Speed
4 Reserved
5 Reserved
6
: Block Descriptor Length
In a MODE SENSE command, the Mode Data Length indicates the total amount of
data available to be transferred. In a MODE SELECT command, this field is 0.
For Ultrium 1 devices, the Medium Type field is not used and must be set to 0.
For Ultrium 2 and later devices, the Medium Type field on Mode Sense data is set
to 00h when no media is loaded, to 18h when Ultrium 1 media is loaded, to 28h
when Ultrium 2 media is loaded, to 38h when Ultrium 3 media is loaded, and to
48h when Ultrium 4 media is loaded..
When the WORM media is loaded, the Medium Type field is logically or’ed with
0x04. Table 95 lists the Medium Type values.
Table 95. Media Type Values
Generation Drive
Medium Loaded 1 2 3 4
No media 00h 00h 00h 00h
Ultrium 1 Data 00h 18h 18h 18h
Ultrium 2 Data N/A 28h 28h 28h
Ultrium 3 Data N/A N/A 38h 38h
Ultrium 3 WORM N/A N/A 3Ch 3Ch

Table 95. Media Type Values (continued)
Generation Drive
Medium Loaded 1 2 3 4
Ultrium 4 Data N/A N/A N/A 48h
Ultrium 4 WORM N/A N/A N/A 4Ch
The Write Protect field indicates whether the currently loaded tape is write
protected in a MODE SENSE command. It is ignored in a MODE SELECT
command.
Buffered Mode values dictate the behavior of the drive as follows:

v 0h (unbuffered) means that the drive will not report Good status on WRITE
commands until the data blocks are actually written on the medium.
v 1h (buffered) means that the drive may report Good status on WRITE
commands as soon as all the data specified in the WRITE command has been
transferred to the logical unit’s buffer. One or more blocks may be buffered prior
to writing the blocks to the medium.
v 2h - 7h (unsupported modes) means that the drive will default to a value of 1h
(buffered).
Speed: 0h (use default speed) (changeable)
Note: Selecting a specific speed is not recommended. The drive is designed to

dynamically select the optimal speed to achieve maximum systemic
performance. This is based on complex criteria including interface
bandwidth, host throughput, data compressibility, etc.
The Block Descriptor Length is 0 if no Block Descriptor is transferred or 8 if a

Block Descriptor is transferred. The Mode parameter value is stored for each
initiator.

Mode Block Descriptor
Table 96. Mode Block Descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0 Density Code
1
: Number of Blocks (0)
3
4 Reserved
5
: Block Length
The Density Code field returns a code identifying the format of the tape currently
loaded in the drive. If there is no tape loaded, the code for the highest capacity
format supported by the drive is returned. See Table 178 on page 187 for details of
supported density codes.
The Number of Blocks field is 0.
The Block Length field indicates the length (in bytes) of each logical block to be
used in subsequent READ, WRITE and VERIFY commands when the Fixed field is
set to 1. (See “READ” on page 154, “VERIFY” on page 255, and “WRITE” on page
256.) A Block Length value of 0 indicates that only variable block transfers are
allowed. The default value is 0. On Parallel SCSI or Fibre Channel devices this
value must be an even number. On SAS devices, block lengths are limited to
multiples of four. If a transfer of a unsupported block size is desired, a variable
length (Fixed field set to 0 in READ, VERIFY, or WRITE command) transfer must
be used.

Mode Page 01h: Read-Write Error Recovery Page
Table 97. Read-Write Error Recovery Page
Bit
Byte 7 6 5 4 3 2 1 0
0 PS (0) Reserved Page Code (01h)
1 Additional Page Length (0Ah)
2 Reserved TB (0) Reserved EER(1) PER DTE(0) DCR(0)
3 Read Retry Limit
4 Reserved
5 Reserved
6 Reserved
7 Reserved
8 Write Retry Limit
9 Reserved
10 Reserved
11 Reserved
On Ultrium 1 drives, a post error (PER) field is set to 0.
On Ultrium 2 and later drives, a post error (PER) field of 1 specifies that the tape
drive will return Check Condition status to report recovered errors. A PER field of
0 specifies that the tape drive will not report errors that are recovered within the
limits established by the error recovery parameters. If this field is 0, the disable
transfer on error (DTE) field must also be set to 0.
A DTE field of 0 specifies that the tape drive will not terminate the transfer for
errors that are recovered within the limits that are established by the read-write
error recovery parameters. On Ultrium drives, the DTE is always set to 0.
The Read Retry Limit is defined in Table 98

Table 98. Read Retry Limit Definition
Value Description
00h - 04h These values will be rounded to 05h
05h Limited error recovery; < 5 seconds.
06h - FEh These values may be rounded (to non-FFh)
FFh Full Recovery Routines allowed (no time limit). (Default)
The Write Retry Limit is defined in Table 99 on page 114

Table 99. Write Retry Limit Definition
Value Description
06h - FEh These values may be rounded (to non-FFh)
FFh Full Recovery Routines allowed (no time limit). (Default)

Mode Page 02h: Disconnect/Reconnect Page
Table 100. Disconnect/Reconnect Mode Page
Bit/Byte 7 6 5 4 3 2 1 0
0 PS SPF (0) Page Code (02h)
1 Page Length (0Eh)
2 Buffer Full Ratio (00h)
3 Buffer Empty Ratio (00h)
4 (MSB)
Bus Inactivity Limit
5 (LSB)
6 (MSB)
Disconnect Time Limit (0000h)
7 (LSB)
8 (MSB)
Connect Time Limit
9 (LSB)
10 (MSB)
Maximum Burst Size
11 (LSB)
12 EMDP(0) Fair Arbitration (0h) DIMM DTDC (0h)
13 Reserved
14 (MSB)
First Burst Size (0000h)
15 (LSB)
The Bus Inactivity Limit field for parallel SCSI and Fibre Channel drives is zero.
For SAS drives this field is reported but is not changeable.
The Connect Time Limit field for parallel SCSI and Fibre Channel drives is zero.
For SAS drives this field is reported but is not changeable.
The Maximum Burst Size field for the Fibre Channel, SAS, and Ultra 320 interfaces
indicates the sequence size that the drive should attempt to use when transferring
data. The value for the Maximum Burst Size field is stored for each initiator.
For Fibre Channel and SAS attached drives

v Any value is allowed and ignored.
For Ultra 320 attached drives:

v If information unit transfers are disabled the Maximum Burst Size field indicates
the maximum amount of data that the SCSI target port shall transfer during a
DATA phase before doing a physical disconnect if the SCSI initiator port has
granted the physical disconnect privilege.
v If information unit transfer are enabled the Maximum Burst Size field indicates
the maximum amount of data that the SCSI target port shall transfer in a single
SPI data information unit.
v The maximum burst size is expressed in increments of 512 bytes (e.g., a value of
one means 512 bytes, two means 1 024 bytes, etc.). A value of zero indicates
there is no limit on the amount of data transferred per burst.
On Ultra 320 attached drives, the DIMM bit is set to one. On all other drives the
DIMM bit is set to 0.

The Buffer Full Ratio and Buffer Empty Ratio fields will be 0, because buffer
management is controlled by the drive.
The Disconnect Time Limit field is not supported and must be set to 0.
The Data Transfer Disconnect Control field is not supported and must be set to 0.

Mode Page 0Ah: Control Extension Mode Page
The mode page policy for this mode page is shared.
Table 101. Control Extension Mode Page
Bit
Byte
7 6 5 4 3 2 1 0
0 PS SPF (1b) Page Code (0Ah)
2 (MSB)
Page Length (1Ch)
3 (MSB)
4 Reserved TCMOS SCIP IALUAE(0)
5 Reserved Initial Priority (0)
6
Reserved
31
A Timestamp Changeable by Methods Outside this Standard (TCMOS) bit set to

one specifies that the timestamp may be initialized by methods outside the scope
of this standard (for example The LDI Set Timestamp command). A TCMOS bit set
to zero specifies that the timestamp shall not be changed by any method except
those defined by this standard.
A SCSI precedence (SCSIP) bit set to one specifies that the timestamp changed
using a SET TIMESTAMP command takes precedence over methods outside the
scope of this standard (for example The LDI Set Timestamp command). A SCSIP
bit set to zero specifies that methods outside this standard (for example The LDI
Set Timestamp command) may change the timestamp and that the SET
TIMESTAMP command is illegal.

Mode Page 0Fh: Data Compression Mode Page
Table 102. Data Compression Mode Page
Bit
Byte 7 6 5 4 3 2 1 0
0 PS (0) Reserved Page Code (0Fh)
1 Additional Page Length (0Eh)
2 DCE DCC(1) Reserved
3 DDE(1) RED (00b) Reserved
4
: Compression Algorithm (1)
7
8
: Decompression Algorithm (1)
11
12 Reserved
13 Reserved
14 Reserved
15 Reserved
A data compression enable (DCE) field of 1 indicates that data compression is

determined by the state of the Select Data Compression Algorithm field of the
Sequential Access Device Configuration Page (10h). The Sequential Access Device
Configuration Page (10h) need not be present in the same MODE SELECT
command. The drive will examine the current state of the Select Data Compression
Algorithm to determine if compression is to be enabled or disabled. The default for
DCE is 1. If DCE is 0, the drive uses Scheme 2 of the LTO-DC algorithm
(passthrough mode).
Note: The result of this is that the only way to turn compression on is for both the
DCE field of the Data Compression Mode Page (0Fh) and the Select Data
Compression Algorithm field of the Sequential Access Device Configuration
Page (10h) to be set to 1.
A data decompression enable (DDE) field is set to 1 to specify that data

decompression is enabled.
The report exception on the decompression (RED) field is set to 00b and specifies
the response to certain boundaries that the drive detects in the data on the
medium. Table 103 on page 119 describes those responses.

Table 103. Responses to Data Boundaries
Response Sense Key
Prior Data Current Data
(see Notes® 1 and 2)
Uncompressed Compressed unsupported algorithm Medium Error
Uncompressed Compressed supported algorithm None
Compressed supported algorithm Uncompressed None
Compressed supported algorithm Compressed unsupported algorithm Medium Error
Compressed unsupported algorithm Uncompressed None
Compressed unsupported algorithm Compressed supported algorithm None
All other combinations -- None
Notes:
1. None in the Response Sense Key column means that no Check Condition status is returned, given the data
boundary condition and the current value of the Report Exception on the Decompression (RED) field.
2. The appropriate additional sense code (ASC) is specified as follows:
v The drive will return a Check Condition status when data is encountered on a medium (during a read
operation) that the device is unable to decompress. In this table, data boundaries that are marked other than
None in Response Sense Key column will generate Check Condition status with the specified sense key.
v If the application client selects an algorithm that the drive does not support, the drive will return a Check
Condition status. The Sense Key must be set to Illegal Request and the ASC must be set to Invalid Field in
Parameter List. The SELECT DATA COMPRESSION ALGORITHM field in the Device Configuration mode
page will be ignored if a Data Compression mode page with the DCE field set to 1 is also received by the
device in the same MODE SELECT command.
No other fields are changeable.

Mode Page 10h: Sequential Access Device Configuration Page
Table 104. Sequential Access Device Configuration Page
Bit
Byte 7 6 5 4 3 2 1 0
1 Additional Page Length (0Eh)
2 Reserved CAP (0) CAF (0) Active Format (0)
3 Active Partition (0)
4 Write Buffer Full Ratio (0)
5 Read Buffer Empty Ratio (0)
6
: Write Delay Time
7
8 DBR (0) BIS (1) RSmk (0) AVC (0) SOCF (0) RBO (0) REW (0)
9 Gap Size (0)
10 ED Defined (0) EEG (1) SEW (0) SWP (0) Reserved
11
: Buffer Size at Early Warning (0)
13
14 Select Data Compression Algorithm
15 WTRE OIR Rewind on Reset (0) ASOCWP PERSWP PRMWP (0)
(0) (0)
The WRITE DELAY TIME field specifies the maximum time, in 100 ms increments,
that the device server should wait before any buffered data that is to be written, is
forced to the medium after the last buffered WRITE command that did not cause
the object buffer to exceed the write object buffer full ratio.
WARNING
Changing the Write Delay Time may result in adverse performance
The Active Partition field will be 0 because multiple partitions are not supported.
The Change Active Format (CAF) and Active Format fields will be 0 because
changing formats is not supported.
The Write Buffer Full Ratio and Read Buffer Empty Ratio fields will be 0 because
buffer management is done by the drive.
The Data Buffer Recovery (DBR), Report Set Marks (RSmk), Stop On Consecutive
File Marks (SOCF), Recover Buffer Order (RBO), Report Early Warning on Read
(REW), and Synchronize at Early Warning fields must be set to 0 because these
features are not supported.

The Automatic Velocity Control (AVC) field must be set to 0 because velocity
control is managed by the drive.
The Gap Size field must be set to 0 because there is no concept of inter-block gaps
in the format.
The Block Identifiers Supported (BIS) field must be set to 1 because block
identifiers are supported.
The enable EOD generation (EEG) field must be set to 1 because the drive always
generates EOD.
The Buffer Size at Early Warning field will be 0, as this cannot be set.
The default value for the Select Data Compression Algorithm is 1 and indicates
that data compression is enabled if the state of the DCE field of the Data
Compression Mode Page (0Fh) is set to 1. The Data Compression Mode Page (0Fh)
need not be present in the same MODE SELECT command. The drive will examine
the current state of the DCE to determine if compression is to be enabled or
disabled. If Select Data Compression Algorithm is 0, the drive uses Scheme 2 of the
LTO-DC algorithm (passthrough mode).
Note: The result of this is that the only way to turn compression on is for both the
DCE field of the Data Compression Mode Page (0Fh) and the Select Data
Compression Algorithm field of the Sequential Access Device Configuration
Page (10h) to be set to 1.
The WORM Tamper Read Enable (WTRE) field is supported only on Ultrium 3 and
later drives.
The WORM Tamper Read Enable (WTRE) field has no effect on the processing of a
locate, read, space, or verify operation when the drive contains a non-WORM
medium.
Only If Reserved (OIR): Changeable on Ultrium 3 and later drives. Default is zero.
Set to zero and not changeable on Ultrium 1 and Ultrium 2 drives.
The WTRE field specifies how the drive responds to detection of compromised
integrity of a WORM medium when processing a locate, read, space, or verify
operation.
Value Action Taken by Drive

If the drive detects compromised integrity on a WORM medium, it will return CHECK
CONDITION status and set the sense key to MEDIUM ERROR and the additional sense code
00b to WORM MEDIUM - INTEGRITY CHECK. (3/300Dh).
The position of the medium may have changed.

Detection of compromised integrity on a WORM medium shall not affect processing of a task.
01b Note: An application client should set the WTRE bit to 01b only for the recovery of data from
a WORM medium where the integrity of the stored data has been compromised.
If the drive detects compromised integrity on a WORM medium it will return CHECK
CONDITION status and set the sense key to MEDIUM ERROR and the additional sense code
10b to WORM MEDIUM - INTEGRITY CHECK. (3/300Dh).
The position of the medium may have changed.

11b Reserved

The drive sets the WTRE field to 10b following a unit attention condition for a
not-ready-to-ready transition.

Mode Page 18h: Protocol-Specific Logical Unit Control Page
There is one copy of this page for each initiator. This page is defined for
Fibre-Channel-attached devices in Table 105. This page is defined for SAS attached
devices in Table 106.
Table 105. Fibre Channel Logical Unit Control Page
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Length (06h)
3 Reserved EPDC
4
: Reserved
Enable Precise Delivery Control (EPDC), when set to 1b, enables checking of a
Fibre Channel Command Reference Number and ensures that the command
packets are delivered in order.
The Protocol Identifier is set to 0h, which indicates that this is for use with the FCP
protocol.
Table 106. SAS Logical Unit Control Page
Bit/Byte 7 6 5 4 3 2 1 0
0 PS SPF(0)
1 Page Length (06h)
2 Reserved TLR (1) Protocol Identifier (6h)
3 Reserved
4 (MSB) Reserved
7 (LSB)

Mode Page 19h: Protocol Specific Port Control Page
There is one copy of this page per primary port. This page is defined for Fibre
channel drives in Table 107 and for SAS drives in Table 108 on page 125.
Ultra 320 drives provide the short format page shown in Table 109 on page 126 as
well as additional subpages defined in Table 110 on page 127 and “Mode Page
19h[04h]: Report Transfer Capabilities Mode Subpage” on page 129. Note that SPI-5
states for Parallel SCSI devices, “Each SCSI target port shall maintain an
independent set of Port Control mode page parameters for each SCSI initiator
port.“ The Ultra 320 drives only support one set of Port Control mode page
parameters.
Table 107. Fibre Channel Port Control Page
Bit
Byte 7 6 5 4 3 2 1 0
1 Page Length
2 Reserved
3 DTFD (0) PLPB (0) DDIS (0) DLM (0) RHA (0) ALWI (0) DTIPE (0) DTOLI (0)
4
: Reserved
5
6 Reserved RR_TOV Units
7 Resource Recovery Time Out Value (RR_TOV)
8 Reserved Control MCM (01b)
9
: Reserved
10
11 Originator CMRs Per Port (0)
12 Reserved
13 Responder CMRs Per Port (0)
14
: MCM_TOV (0)
15
The Page Length field is returned by Mode Sense commands and should be set by
Mode Select commands on Ultrium 1 drives as 0Eh, and on Ultrium 2 and later
drives as 06h.
The Page Length field returns the number of remaining bytes. On Ultrium 1 drives
this value is 0Eh. On Ultrium 2 and later drives this value is 06h.
The Resource Recovery Time Out Value (RR_TOV) is the minimum amount of time
that the drive will wait for an expected response before implicitly cleaning up the
resources that are related to that initiator. This may, depending on the
circumstances, implicitly log-out the initiator that stopped communicating with the
drive.
Care should be taken when adjusting this value, because a value that is too small
has the potential to cause resources to be discarded prior to the completion of a

class 3 error recovery and to prematurely log-out an initiator. It also has the
potential, when the value is set too large, to cause command time-outs for
non-failing initiators in a multi-initiator environment, if one of the initiators fails.
The Protocol Identifier is set to 0h, which indicates that this is for use with the FCP
protocol.
Resource Recovery Time Out Value (RR_TOV) Units can have the following values:
v 000b (no timer is specified)
v 001b (timer is specified in .001-second units)
v 011b (timer is specified in .1-second units)
v 101b (timer is specified in 10-second units)
RR_TOV Value can be between 0 and FFh.
The following conditions will round the RR_TOV. If the value is rounded, a
Recovered Error, Mode Parameters Rounded Check Condition is presented.
v The value of RR_TOV that is determined by the RR_TOV Units and RR_TOV
Value fields is less than the minimum supported value (RR_TOV set to
Minimum Value)
v The value of RR_TOV that is determined by the RR_TOV Units and RR_TOV
Value fields is greater than the maximum supported value (RR_TOV set to
Maximum Value)
v The RR_TOV Units is an unsupported value (RR_TOV set to Default Value)
Note that when the RR_TOV value is returned from the drive, it may be returned
using different RR_TOV Units than were used to set the value in a previous Mode
Select command.
Table 108. SAS Port Control Page
Bit/Byte 7 6 5 4 3 2 1 0
1 Page Length (06h)
Ready
LED
Meaning
(0)
3 Reserved
4 (MSB)
I_T Nexus Loss Time (07D0h)
5 (LSB)
6 (MSB)
Initiator Response Timeout (07D0h)
7 (LSB)

Mode Page 19h: Protocol Specific Port Control Page Short
Format (Ultra 320 Attached Drives Only)
The Protocol Specific Port Control Page short format for Ultra 320 drives is defined
in Table 109.
Table 109. Protocol Specific Port Control Page short format (U320 only)
Bit/Byte 7 6 5 4 3 2 1 0
1 Page Length (06h)
2 Reserved Protocol Identified (1h)
3 Reserved
4 (MSB)
Synchronous Transfer Timeout (0000h)
5 (LSB)
6 Reserved
7 Reserved

Mode Page 19h[03h]: Negotiated Settings Mode Subpage
The Negotiated Settings mode subpage is used to report the negotiated settings of
the drives port and is defined in Table 110.
Table 110. Negotiated Settings mode subpage (U320 drives only)
Bit/Byte 7 6 5 4 3 2 1 0
2 (MSB)
Page Length (0Ah)
3 (LSB)
4 Reserved
6 Transfer Period Factor
7 Reserved
8 REQ/ACK Offset
9 Transfer Width Exponent
Protocol Options Bits
10 Rsvd RTI RD_STRM WR_ HOLD_ QAS_ DT_REQ IU_REQ
FLOW MCS REQ
Sent Received
11 Reserved Transceived Mode (10b)
PCOMP_EN PCOMP_EN
12 Reserved
13 Reserved
The Transfer Period Factor field indicates the negotiated transfer period factor for
the current I_T nexus. See Table 111 on page 128.

Table 111. Transfer Period Factor
Value Description Transfer Rate
00h-06h Reserved N/A
07h Transfer period equals 3.125 ns Fast-320
0Ah Transfer period equals 25 ns Fast-40
0Bh Transfer period equals 30.3 ns Fast-40
0Ch Transfer period equals 50 ns Fast-20
0Dh - 18h Transfer period equals the TRANSFER PERIOD FACTOR x 4 Fast-20
19h - 31h Transfer period equals the TRANSFER PERIOD FACTOR x 4 Fast-10
32h - FFh Transfer period equals the TRANSFER PERIOD FACTOR x 4 Fast-5
The REQ/ACK Offset field indicates the negotiated REQ/ACK offset for the
current I_T nexus
The Transfer Width Exponent field indicates the negotiated transfer width
exponent for the current I_T nexus. See Table 112
Table 112. Transfer Width Exponent
Value Description
00h Specifies 8 bit data bus (i.e., narrow transfer agreement).
01h Specifies 16 bit data bus (i.e., wide transfer agreement).
02h Obsolete
03h-FFh Reserved
The Protocol Options Bits field contain the negotiated protocol options for the
current I_T nexus.
The Received PCOMP_EN bit contains the value of the PCOMP_EN bit received
by the SCSI target port for the current I_T nexus.
The Sent PCOMP_EN bit contains the value of the PCOMP_EN bit sent by the
SCSI target port for the current I_T nexus.

Mode Page 19h[04h]: Report Transfer Capabilities Mode
Subpage
The Report Transfer Capabilities mode subpage is used to report transfer
capabilities of the drives port and is defined in Table 113.
Table 113. Report Transfer Capabilities mode subpage (U320 drives only)
Bit/Byte 7 6 5 4 3 2 1 0
2 (MSB)
Page Length (0Ah)
3 (LSB)
4 Reserved
6 Minimum Transfer Period Factor (06h)
7 Reserved
8 Maximum REQ/ACK Offset
9 Maximum Transfer Width Exponent (01h)
Protocol Options Bits Supported
10 PCOMP RD_STRM WR_FLOW HOLD_MCS QAS_REQ DT_REQ
RTI (0) IU_REQ (1)
_EN(1) (0) (0) (0) (0) (1)
11 Reserved
12 Reserved
13 Reserved
The Minimum Transfer Period Factor field shall be set to the smallest value of the
transfer period factor supported by the SCSI target port.
The Maximum REQ/ACK Offset shall be set to the largest value of the REQ/ACK
offset supported by the SCSI target port.
The Maximum Transfer Width Exponent shall be set to the largest value of the
transfer width exponent supported by the SCSI target port.
The SCSI target port shall set the bits in the Protocol Options Bits Supported field
to indicate the protocol options supported by the SCSI target port.

Mode Page 1Ch: Information Exceptions Mode Page
Table 114. Information Exceptions Mode Page
Bit
Byte 7 6 5 4 3 2 1 0
0 PS (0) Reserved Page Code (1Ch)
1 Page Length (0A)
2 Perf (0) Reserved DExcpt Test Reserved LogErr(0)
3 Reserved MRIE (3)
4
: Interval Timer (0)
7
8
: Report Count (0)
11
The Information Exceptions mode page is used to control Exception Reporting by

using the TapeAlert log page.
A disable exception control (DExcpt) field of 0 indicates that reporting for failure
prediction threshold exceeded will be enabled. The method for reporting the failure
prediction threshold exceeded when the DExcpt field is set to 0 is determined from
the MRIE field. A DExcpt field of 1 indicates that the target will disable reporting
of the failure prediction threshold exceeded. The default value for DExcpt is 0.
If the Test field is set to 1, the next command will fail, the Sense Key will be set to
Unit Attention, and the ASC/ASCQ will be set to Failure Prediction Threshold
Exceeded - False (5DFF). If the Test field is set to 0, the next command is processed
normally. The default for Test is 0.
The Perf and LogErr fields will be 0. These features are not supported.
The MRIE field must be set to 3 (Conditionally generate recovered error). This
method instructs the drive to report informational exception conditions (if the
reporting of recovered errors is allowed) by returning a Check Condition status. If
the Test field is set to 0, the status may be returned after the informational
exception condition occurs on any command for which Good status would have
been returned. If the Test field is set to 1, the status will be returned on the next
command that is normally capable of returning an informational exception
condition when the Test field is set to 0. The Sense Key must be set to Recovered
Error and the Additional Sense Code will indicate the cause of the informational
exception condition. This will be Failure Prediction Threshold Exceeded (5D00) if
the Test field is set to 0 (true error), and Failure Prediction Threshold Exceeded
-False (5DFF) if the Test field is set to 1 (test).
The command that returns the Check Condition status for the informational
exception will complete without error before any informational exception condition
may be reported. The Interval Timer and Report Count must be set to 0. These
fields are not supported.

Mode Page 1Dh: Medium Configuration Mode Page
The Medium Configuration Mode Page (see Table 115) specifies any special
considerations the drive will use when processing commands when there is a
WORM medium loaded in the drive. This page is supported only on Ultrium 3
and later drives. This page is returned when requested by a Mode Sense command
indicating either this page or all pages (3Fh).
Table 115. Medium Configuration Mode Page
Bit
Byte 7 6 5 4 3 2 1 0
0 PS (0) SPF(0) Page Code (1Dh)
1 Page Length (1Eh)
2 Reserved WORMM
3 Reserved
4 Worm Mode Label Restrictions (01h)
5 Worm Mode Filemark Restrictions (02h)
6-31 Reserved
The WORMM bit is set to 1 when the drive is operating in WORM mode. The
WORMM bit is set to 0 when the drive is not operating in WORM mode. If a
Mode Select command is processed that attempts to change the setting of the
WORMM bit, the drive returns a Check Condition status, with the sense key set to
Illegal Request, and the addition sense code set to Invalid Field In Parameter List.
The Worm Model Label Restrictions field specifies the restrictions against
overwriting format labels when operating in WORM mode (see Table 116).
A series of filemarks with no interleaved logical blocks immediately preceding

EOD is treated as a filemark sequence and controlled by the Worm Mode
Filemarks Restrictions field.
Table 116. Worm Mode Label Restrictions field values
Worm Mode
Description
Label Restrictions
00h The drive does not allow any logical blocks to be overwritten. (Not Supported)
The drive allows a tape header to be overwritten. The tape header is defined as 0, 1,
01h or 2 logical blocks followed by nothing except 0 to n Filemarks and EOD. This must
be overwritten from BOP.
02h The drive allows all format labels to be overwritten. (Not Supported)
03h - FFh Reserved
The Worm Mode Filemarks Restrictions field specifies the restrictions against
overwriting a series of filemarks immediately preceding EOD when operating in
WORM mode (see Table 117 on page 132). This field controls only the overwriting
of a series of filemarks with no interleaved logical blocks immediately preceding
EOD.

Table 117. Worm Mode Filemarks Restrictions field values
Worm Mode
Description
Label Restrictions
00h - 01h Reserved
The drive allows any number of filemarks immediately preceding EOD to be
02h
overwritten, except the filemark closest to BOP.
The drive allows any number of filemarks immediately preceding EOD to be
03h
overwritten. (Not Supported)
04h - FFh Reserved

Mode Page 24h: Vendor-Specific Control Mode Page
The Vendor-Specific Control Mode Page provides control over undocumented test
options. Modifying these values could have adverse effects on the drives operation.
As such, this page should never be used in a Mode Select operation. A Mode Sense
of this page may be used to determine if the drive supports encryption. The
Vendor-Specific Control Mode Page is defined in Table 118.
Table 118. Vendor-Specific Control Mode Page
Bit
Byte 7 6 5 4 3 2 1 0
0 PS (0) SPF(0) Page Code (24h)
1 Page Length (06h)
2 Vendor-Specific Control
3 Vendor-Specific Setting
4
Vendor-Reserved
6
7 Vendor-Reserved Encr_E Vendor-Reserved Encr_C
The encryption enabled (Encr_E) bit set to zero indicates that encryption is not
enabled in the drive. Data written will be written in clear text. The Encr_E bit set
to one indicates that encryption is enabled in the drive.
Note: There are multiple encryption methods, Application Managed Encryption

(AME), System Managed Encryption (SME), and Library Managed
Encryption (LME). When the drive is configured for SME or LME
encryption is transparent to the application and the Security Protocol In
command will report no encryption support.
The encryption capable (Encr_C) bit set to zero indicates that the drive hardware
does not support encryption. The Encr_C bit set to one indicates the drive
hardware supports encryption but does not indicate if encryption is currently
enabled.

Mode Page 2Fh: Behavior Configuration Mode Page
There is one copy of this page for the drive.
Ultrium drive support for the fields in this page vary by generation and/or code
level. To discover if a field can be modified, issue a Mode Sense with a PC field of
01 to see if the field is changeable. This will also return the Page Length, which
must be examined to determined the length of this page, because it is expected to
increase as additional Behavior configurations are added to subsequent code levels.
WARNING: This page changes the normal behavior of the drive. Some settings
effect error reporting and may even cause the drive to violate SCSI standards. Care
should be taken to ensure that the behavior change is understood and what its
effects on the system will be.
Table 119. Behavior Configuration Mode Page
Bit
Byte 7 6 5 4 3 2 1 0
0 PS Reserved Page Code (2Fh)
1 Page Length (n-1)
2 Fence Behavior
3 Clean Behavior
4 WORM Behavior
5 Sense Data Behavior
Reporting Behavior
6
Rsvd Rsvd Rsvd Rsvd Rsvd CCDM DDEOR CLNCHK
Firmware Update Behavior
7
Rsvd Rsvd Rsvd Rsvd Rsvd Rsvd Rsvd DFMRDL
Unload On Error Behavior
8
Reserved UOE-C UOE-F UOE-D
.
.
.
n Reserved for Future Use

Fence Behavior (Byte 2)
This field defines drive behavior in situations deemed dangerous for either media
or data on the media.
Table 120. Fence Behavior Selection Values
Description Value
Normal Operation (default) 00h
Mid-Tape Recovery (MTR) Fence Only
Panic Fence Feature Enabled 01h
Reserved 02h - FFh
Normal Operation (00h): The behavior of the drive after a Panic or Exception
when in normal operation is defined below).
When the drive comes up after a Panic or Exception and no cartridge is detected,
the drive will respond to SCSI commands as follows (Steps 1 and 2 are returned in
response to any check condition eligible command):
1. 6/2900 sense is returned.
2. 2/3e00 is returned during POST.
3. When POST completes, enter normal operation.
When the drive comes up and detects a cartridge, it requires Mid-Tape Recovery
(MTR) and will respond to SCSI commands as follows (Steps 1 through 5 are
returned in response to any check condition eligible command):
1. 6/2900 is returned.
3. 2/0400 is returned during Mid-Tape Recovery/Unload.
4. 2/0401 is returned during Mid-Tape Recovery/Load.
5. 6/2800 is returned after cartridge is loaded.
6. Enter into the MTR Fence State.
a. TUR commands will return GOOD status.
b. Return 5/2C00 Sense for all medium access commands.
c. Exit MTR Fence State when an explicit positioning command completes
successfully (that is, Locate, Rewind, Load).
Panic Fence Feature Enabled (01h): The behavior of the drive after the Panic or
Exception when the Panic Fence Feature is enabled is defined below.
When the drive comes up after a Panic of Exception and no cartridge is detected,
the drive will respond to SCSI commands as follows (Steps 1 and 2 are returned in
response to any check condition eligible command):
3. When POST completes, enter into the Panic Fence state. In the Panic Fence
state:
a. SCSI commands other than RSNS/INQ/RLUNs/Read Buffer/TUR are
rejected with 5/2904 sense, indicating Panic Fence state.
b. TUR commands will return 5/2904 sense.
c. Load of a cartridge through any means is not allowed.
d. SCSI Read Buffer to read dump data is accepted at any time.

e. Once dump has been read, go to normal mode.
When the drive comes up and detects Mid-Tape Recovery (MTR), it will respond
to SCSI commands as follows (steps 1 through 5 are returned in response to any
check condition eligible command):
3. 2/0400 is returned during Mid-Tape Recovery/Unload.
4. 2/0401 is returned during Mid-Tape Recovery/Load.
5. 6/2800 is returned after cartridge is loaded.
6. Enter into the Panic Fence state.
a. SCSI commands other than RSNS/INQ/RLUNs/Read Buffer/TUR/Load
with the load bit set to 0 are rejected with 5/2904 sense, indicating Panic
Fence state.
b. TUR commands will return GOOD status while tape is loaded.
c. TUR commands will return 5/2904 sense after the tape is unloaded.
d. Unload command through SCSI/LDI/Button can be executed anytime.
e. Load of a cartridge through any means is not allowed.
f. SCSI Read Buffer to read dump data is accepted at any time.
g. Once dump has been read, if the tape is still loaded, transition to MTR
Fence state. If tape is still not loaded, exit from Panic Fence state and go to
normal mode.
7. MTR Fence state.
a. TUR commands will return GOOD status.
b. Return 5/2C00 Sense for all medium access commands.
c. Exit MTR Fence State when an explicit positioning command completes
successfully (that is, Locate, Rewind, Load).

Clean Behavior (Byte 3)
This field defines the behavior of the drive related to cleaning.
Description Value
Normal Operation (default) 00h
Periodic Clean Notification Enabled 01h
Reserved 02h - FFh
Note: On eServer enabled drives (i.e., The Product ID returned in Standard

Inquiry is "HH LTO Gen x" where "x" is a number) that are attached to AIX,
the Clean Behavior gets modified by the SCDD to use Periodic Clean
Notification Enabled (01h).
Periodic Clean Notification Enabled (01h): The drive shall monitor the number of
write and read datasets since last cleaning. When the number of write/read
datasets exceeds the criteria the drive will put itself in a clean notification needed
state. The criteria used is subject to change, but the current criteria is listed in
Table 121.
Table 121. Periodic Clean Notification Usage Criteria
Equivalent Full File
Meters of tape
Generation Drive Data Sets Processed Passes
pulled across head
(Approximate)
Ultrium 1 10,000,000 38 N/A
Ultrium 2 10,000,000 19 N/A
Ultrium 3 10,000,000 38 N/A
Ultrium 4 12,000,000 23 5,000,000
When entering this state the drive shall set the Drive Status Flags 1, Byte 7, Bit 6:
Drive Clean Required as defined in the LDI and the drive shall set TapeAlert flag
21.
This drive will remain in this state until a successful cleaning cycle or a Power On
Reset occurs. (e.g. If there is another data cartridge inserted without the drive
having been cleaned, the drive will function as normal. Once the TapeAlert is
reported the first time it will be cleared following normal TapeAlert rules. The
Drive Status Flags will function as defined in the LDI.)

WORM Behavior (Byte 4)
This field defines the behavior of the drive related to WORM.
Table 122. WORM Behavior
Description Value
Normal Operation (default - reset to this value on
00h
Medium Removal)
Data Cartridge Emulates WORM (Ultrium 3 and later) 01h
Reserved 02h - FFh
Data Cartridge Emulates WORM (01h): Treat Data Cartridge like a WORM
cartridge. When a data cartridge is inserted into the drive the drive will report that
it is WORM media and the drive will behave like the media is WORM. It will
follow all overwrite rules. This Emulate mode will be cleared after cartridge
unload. There will be no emulation of the WORM tampering responses. Sense code
of 300Dh is never reported in WORM Emulation mode. WTRE can be set in the
Sequential Access Device Configuration Mode page (10h), this does not affect drive
behavior in emulation mode.

Sense Data Behavior (Byte 5)
Table 123. Sense Data Behavior
Description Value
Use 35-byte request Sense data
(Not supported on Ultrium 3 and later Drives) 00h
(Default on Ultrium 2 RoHS Drives)
Use 96-byte Request Sense data
01h
(Not supported on Ultrium 1 and Ultrium 2 non-RoHS Drives)
Reserved for future use 02h - FFh
Note: Regardless of this setting, when the sense data is associated with an Illegal Length read, an 18-byte Request
Sense data may be used.

Reporting Behavior (Byte 6)
This byte defines behaviors related to reporting conditions across the primary
interface.
A Clean Check (CLNCHK) bit set to one specifies that a Check Condition, Sense
Key 0h ASC/ASCQ 8282h (Drive Requires Cleaning) will be returned after a
Rewind, Space, Locate, or Unload when cleaning is required. A CLNCHK bit set to
zero does not specify that a Check Condition will be returned when cleaning is
required. The default value of this bit is zero (i.e. No check condition is returned).
A Disable Deferred Error On Rewind (DDEOR) bit set to one specifies that no
deferred error will be reported to a rewind command. A DDEOR bit set to zero
does not specify that no deferred error will be reported on rewind. The default
value of this bit is zero.

the Disable Deferred Error On Rewind (DDEOR) bit gets set to one by the
SCDD.
A Check Condition for Dead Media (CCDM) bit set to one specifies that a Check
Condition, Sense Key = 1h, ASC/ASCQ 8252h (Degraded Media) will be returned
after a Rewind command when the criteria are met to set the Dead Media flag in
Log Page 3Ch. A CCDM bit set to zero does not specify that a Check Condition
will be returned when the Dead Media criteria are met. The default value of this
bit is zero (default is one if manufactured as an eServer drive - i.e., The Product ID
returned in Standard Inquiry is "HH LTO Gen x" where "x" is a number).

Firmware Update Behavior (Byte 7)
This byte defines behaviors related to updating drive firmware.
A Disable Field Microcode Replacement Down Level (DFMRDL) bit set to one
specifies that the drive will not accept downlevel firmware via an FMR tape. A
DFMRDL bit set to zero does not specify that the drive will not accept downlevel
firmware via an FMR tape. This bit does not effect code downloads by means
other than an FMR tape (e.g. Write Buffer over SCSI interface). The default value of
this bit is zero (i.e. Firmware can be down leveled by FMR tape).

the Disable Field Microcode Replacement down Level (DFMRDL) bit is set
to one by the SCDD.

Unload On Error Behavior (Byte 8)
This byte defines auto unload behaviors of different types of cartridges when the
following errors occur:
v Bad Media causes a mount failure;
v FMR cartridge is loaded when it has not been requested;
v Expired cleaner cartridge is loaded; or
v An invalid or unsupported cleaner cartridge is loaded.
Settings specified over a library interface (i.e. LDI or ADI) take precedence over
these settings, but do not modify these values.
Table 124 defines the Unload on Error Behavior fields. Currently the only values
allowed are 00h or 15h, but we anticipate allowing more flexibility in the future.
Table 124. Unload On Error Behavior Definition
Value Field Behavior
Unload On Error - Data (UOE-D)
00b Unload on Error - FMR (UOE-F) No exceptional behavior specified
Unload On Error - Cleaner (UOE-C)
01b Unload on Error - FMR (UOE-F) Do not auto-eject on error
10b Unload on Error - FMR (UOE-F) Reserved
11b Unload on Error - FMR (UOE-F) Reserved

Mode Page 3Dh: LEOT Mode Page (LTO 1 Only)
The LEOT Mode Page is used to change the capacity of the loaded volume and is
defined inTable 125.
Table 125. LEOT Mode Page (Ultrium 1 Only)
Bit
Byte 7 6 5 4 3 2 1 0
0 PS (0) SPF(0) Page Code (3Dh)
1 Reserved
2
3 MSB
Number of Wraps
4 LSB

PERSISTENT RESERVE IN
The PERSISTENT RESERVE IN command is used for reservation management to
show what types of Reservations and Reservation Keys exist.
Table 126. PERSISTENT RESERVE IN Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (5Eh)
2 Reserved
3 Reserved
4 Reserved
5 Reserved
6 Reserved
7
: Allocation Length
8
9 Control
Service Action may have one of the following values:

v 00000b (reads all registered Reservation Keys)
v 00001b (reads all current persistent reservations)
v 00010b (Returns capability information)
v 00011b (Reads complete information about all registrations and the persistent
reservations, if any)
Allocation Length is set to the maximum number of bytes to be transferred.
The PERSISTENT RESERVE IN parameter data for Read Capabilities is defined in

Table 127.
Table 127. PERSISTENT RESERVE IN Parameter Data for Read Capabilities
7 6 5 4 3 2 1 0
0 MSB Length (0008h)
1 LSB
2 Reserved CRH (1) SIP_C (1) ATP_C (1) Reserved PTPL_C (0)
3 TMV (1) Reserved PTPL_A (0)
Persistent Reservation Type Mask (bytes 4-5)
WR_EX_AR EX_AC_RO WR_EX_RO
4 Reserved EX_AC (1) Reserved WR_EX (0) Reserved
(0) (1) (0)
EX_AC_AR
5 Reserved
(0)
6
Reserved
7

Compatible Reservation Handling (CRH): Set to one.
Specify Initiator Ports Capable (SIP_C): Set to one.
All Target Ports Capable (ATP_C): Set to one.
Persist Through Power Loss Capable (PTPL_C): Set to zero.
Type Mask Valid (TMV): Set to one.
Persist Through Power Loss Activated (PTPL_A): Set to zero.
Write Exclusive – All Registrants (WR_EX_AR): Set to zero.
Exclusive Access – Registrants Only (EX_AC_RO): Set to one.
Write Exclusive – Registrants Only (WR_EX_RO): Set to zero.
Exclusive Access (EX_AC): Set to one.
Write Exclusive (WR_EX): Set to zero.
Exclusive Access – All Registrants (EX_AC_AR): Set to zero.
The PERSISTENT RESERVE IN parameter data for Read Full Status is defined in
Table 128.
Table 128. PERSISTENT RESERVE IN Parameter Data for Read Full Status
7 6 5 4 3 2 1 0
0 MSB
PRGeneration
3 LSB
4 MSB
Additional Length (n-7)
7 LSB
Full status descriptors
8
First full status descriptor (see Table 129 on page 146)
.
.
.
Last full status descriptor (see Table 129 on page 146)

n
The format of the full status descriptors is shown in Table 129 on page 146.

Table 129. PERSISTENT RESERVE IN Full Status descriptor format
7 6 5 4 3 2 1 0
0 MSB
Reservation Key
7 LSB
8 MSB
Reserved
11
12 Reserved ALL_TG_PT R_HOLDER
13 SCOPE TYPE
14
Reserved
17
18 MSB
Relative Target Port Identifier
19 LSB
20 MSB
Additional Descriptor Length (n-23)
23 LSB
24
TransportID
n
Reservation Holder (R_HOLDER).

All Target Ports (ALL_TG_PT).
The PERSISTENT RESERVE IN parameter data for Read Keys is defined in

Table 130.
Table 130. PERSISTENT RESERVE IN Parameter Data for Read Keys
Bit
Byte 7 6 5 4 3 2 1 0
0
: Generation
3
4
: Additional Length
7
8
: First Reservation Key
15
16
: Additional Reservation Keys
Generation is a counter for PERSISTENT RESERVE OUT command requests.
Additional Length is a count of the number of bytes in the Reservation Key list.

For Additional Reservation Keys, a maximum of one reservation key per initiator
is supported.
The PERSISTENT RESERVE IN parameter data for Read Reservations is defined in

Table 131.
Table 131. PERSISTENT RESERVE IN Parameter Data for Read Reservations
Bit
Byte 7 6 5 4 3 2 1 0
0
: Generation
3
4
: Additional Length
7
8
: Read Reservations Descriptor
Generation is a counter for PERSISTENT RESERVE OUT command requests.
Additional Length is a count of the number of bytes in the Reservation Key list.
Reservation Descriptors are defined in Table 132.
The PERSISTENT RESERVE IN Read Reservations Descriptor is defined in

Table 132.
Table 132. PERSISTENT RESERVE IN Read Reservations Descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0
: Reservation Key
7
8
: Scope-specific address (0)
11
12 Reserved
13 Scope (0h) Type
14
: Extent Length (0)
15

A Scope value of 0h indicates that the persistent reservation applies to the entire
logical unit.
Type may have one of the following values:

v 3h means Exclusive Access
v 6h means Exclusive Access, Registrants only

PERSISTENT RESERVE OUT
The PERSISTENT RESERVE OUT command is used for reservation management to
allow different types of Reservations and Reservation Keys to be created or
removed.
Table 133. PERSISTENT RESERVE OUT Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (5Fh)
2 Scope (0) Type
3 Reserved
4 Reserved
5 Reserved
6 Reserved
7
: Parameter List Length (18h)
8
9 Control
Table 134 contains the values for Service Action field. For additional information
about the descriptions of each service action code, refer to the SCSI Primary
Commands-3 (SPC-3) manual.
Table 134. Values for Service Action Codes in PERSISTENT RESERVE OUT Command
PERSISTENT
Code Name Description RESERVE Generation
Field Incremented
Registers a reservation key with the device
00h REGISTER Yes
server or unregisters a reservation key.
Creates a persistent reservation that has a
01h RESERVE No
specified SCOPE and TYPE.
02h RELEASE Releases the selected persistent reservation. No
Clears all reservation keys (for example,
03h CLEAR Yes
registrations) and all persistent reservations.
Preempts persistent reservations or removes
04h PREEMPT Yes
registrations.
Preempts persistent reservations or removes
PREEMPT AND
05h registrations and aborts all tasks for all Yes
ABORT
preempted initiator ports.
REGISTER AND
Registers a reservation key with the device
06h IGNORE EXISTING Yes
server or unregisters a reservation key.
KEY
Register a reservation key for another I_T
REGISTER AND
07h nexus with the device server and move a Yes
MOVE
persistent reservation to that I_T nexus
08h - 1Fh Reserved

The value in the Type field specifies the characteristics of the persistent reservation
that is being established for all data blocks within the logical unit. Table 135
defines the characteristics of the different type values. For each persistent
reservation type, Table 135 lists the value of the code, its name, the type of drive
support, and a description of the drive support.
Table 135 contains the values for the Type field.

Table 135. Persistent Reservation Type Codes
Drive
Description of Drive Support
Code Name Support
(see Note 2)
(see Note 1)
0h N/S Not Supported
Write
Exclusive
2h N/S Obsolete
Reads Exclusive: Any task from any initiator port other than the
initiator port that holds the persistent reservation that requests a
transfer from the storage medium or cache of the logical unit to the
initiator port will be terminated with RESERVATION CONFLICT
status.
Writes Exclusive: Any task from any initiator port other than the
Exclusive initiator port that holds the persistent reservation that requests a
3h 1,2
Access transfer from the initiator port to the storage medium or cache of the
logical unit will be terminated with RESERVATION CONFLICT
status.
Persistent Reservation Holder: The initiator port that delivered the
PERSISTENT RESERVE OUT command with RESERVE, PREEMPT, or
PREEMPT AND ABORT service action as identified by its registered
reservation key.
Write
Exclusive -
Registrants
Only
Reads Exclusive: A task that requests a transfer from the storage
medium or cache of the logical unit to an initiator port that is not
currently registered with the device server will be terminated with
RESERVATION CONFLICT status.
Exclusive Writes Exclusive: A task that requests a transfer to the storage
Access - medium or cache of the logical unit from an initiator port that is not
6h 1,2
Registrants currently registered with the device server will be terminated with
Only RESERVATION CONFLICT status.
Persistent Reservation Holder: The initiator port that delivered the
PERSISTENT RESERVE OUT command with RESERVE, PREEMPT, or
PREEMPT AND ABORT service action as identified by its registered
reservation key.
Write
Exclusive -
All
Registrants
Exclusive
8h Access - All N/S Not Supported
Registrants
9h - Fh Reserved

Table 135. Persistent Reservation Type Codes (continued)
Drive
Description of Drive Support
Code Name Support
(see Note 2)
(see Note 1)
Notes:
1. Drive Support is categorized as follows:
1 = Generation 1
2 = Generation 2
N/S = not supported
2. The Description of Drive Support column is divided into three categories:
A definition of the required handling for read operations.
A definition of the required handling for write operations.
A definition of the persistent reservation holder (for more information, refer to the SCSI Primary Commands-3
(SPC-3) manual.
The PERSISTENT RESERVE OUT parameter list for all service actions except
REGISTER AND MOVE (07h) is defined in Table 136.
Table 136. PERSISTENT RESERVE OUT Parameter List
Bit
Byte 7 6 5 4 3 2 1 0
0
: Reservation Key
7
8
: Service Action Reservation Key
15
16
: Scope-specific Address (0)
19
20 Reserved APTPL (0)
21 Reserved
22
: Obsolete (0)
23
Any value is allowed for the Reservation Key and the Service Action Reservation
Key.
The value for Activate Persist Through Power Loss (APTPL) will be 0.
The PERSISTENT RESERVE OUT parameter list for REGISTER AND MOVE (07h)
is defined in Table 137 on page 152.

Table 137. PERSISTENT RESERVE OUT command with REGISTER AND MOVE service action parameter list
Bit
Byte 7 6 5 4 3 2 1 0
0 MSB Reservation Key
7 LSB
8 MSB Service Action Reservation Key
15 LSB
16 Reserved
17 Reserved UNREG APTPL (0)
18 MSB Relative Target Port Identifier
19 LSB
20 MSB TransportID Parameter Data Length (n-23)
23 LSB
24 TransportID
n

PREVENT/ALLOW MEDIUM REMOVAL
The PREVENT/ALLOW MEDIUM REMOVAL command is used to prevent
accidental removal of the medium while it is required by an initiator.
Table 138. PREVENT/ALLOW MEDIUM REMOVAL Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (1Eh)
1 Obsolete Reserved
2 Reserved
3 Reserved
4 Reserved Prevent
5 Control
If the Prevent field is set, then eject requests from the front panel are ignored and
Unload commands give Check Condition status. The Sense Key is set to Illegal
Request and the ASC/ASCQ to Medium Removal Prevented (5302).
All initiators that have prevented medium removal must enable it before the
medium can be removed from the drive.

READ
The READ command causes data to be transferred from the tape medium to the
initiator.
Table 139. READ Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved SILI Fixed
2
: Transfer Length
4
5 Control
If the Fixed field is set to 0 and Transfer Length is not 0, then a single block of the
length in Transfer Length is to be transferred. If the next block on tape is of this
length or shorter, then it is transferred to the initiator. If the next block is longer
than this length, then only the length requested is returned. A Check Condition for
incorrect length is returned, and the logical position is set after the block. If the
length of the block was the same as the Transfer Length field, then Good status is
returned.
If the Suppress Incorrect Length Indicator (SILI) field is 1 and the Fixed field is 0,
the drive will do one of the following:
v Report Check Condition status for an incorrect length condition only if the
overlength condition exists and the BLOCK LENGTH field in the mode
parameter block descriptor is non-zero (see clause 8.3 in the SCSI-3 Stream
Commands (SSC)).
v Not report Check Condition status if the only error is the underlength condition,
or if the only error is the overlength condition and the BLOCK LENGTH field of
the mode parameters block descriptor is 0.
If the SILI field is 0 and an incorrect length block is read, Check Condition status
will be returned. The ILI and VALID fields must be set to 1 in the sense data and
the Additional Sense Code must be set to NO ADDITIONAL SENSE
INFORMATION. Upon termination, the logical position will be after the incorrect
length block (end-of-partition side). If the Fixed field is 1, the INFORMATION field
must be set to the requested transfer length minus the actual number of blocks
read (not including the incorrect length block). If the Fixed field is 0, the
INFORMATION field must be set to the requested transfer length minus the actual
block length.
If the Fixed field is set to 1, the Block Length (see “Mode Block Descriptor” on
page 112) is set to 0, and the Transfer Length field is not 0, Check Condition status
is returned with Illegal Field in CDB (5/2400h).
If the Fixed field is set to 1, the Transfer Length field is not 0, and the Suppress
Illegal Length Indicator (SILI) field is set to 0, then a sequence of blocks of the
currently configured block length is to be returned, the number of blocks being
indicated in the Transfer Length field. If there is a sequence of blocks of this length
on the tape, they are returned to the initiator with Good status. If a block that is
longer than the configured length is encountered before the sequence is complete,
the blocks up to that block are returned, followed by the configured length from
the record that was too long and Check Condition status. If a block that is shorter
than the configured length is encountered before the sequence is complete, the
blocks up to that block are returned, followed by all of that block and Check
Condition status. The current position is set after the last block that was returned
or partially returned.
If the Transfer Length field is 0, and if the Suppress Illegal Length Indicator and
the Fixed field are not both set, then Good status is returned and no action is taken
in the drive.
If Suppress Illegal Length Indicator (SILI) field is set and the Fixed field is set, then
Check Condition status is returned. The Sense Key is set to Illegal Request and the
ASC/ASCQ is set to Invalid Field in CDB (2400).

READ ATTRIBUTE
The READ ATTRIBUTE command allows an application client to read attribute
values from medium auxiliary memory.
Table 140. READ ATTRIBUTE Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Reserved (0) Service Action
2
: Reserved
3
4 Reserved
5 Volume Number (0)
6 Reserved
7 Partition Number (0)
8
: First Attribute ID
9
10
: Allocation Length
13
14 Reserved
15 Control
If the medium auxiliary memory is not accessible because there is no medium

present, the READ ATTRIBUTE command will be terminated with a Check
Condition status. The Sense Key must be set to Not Ready and the Additional
Sense Code must be set to Medium Not Present (3A00h).
If the medium auxiliary memory is not accessible but the medium is present, the
READ ATTRIBUTE command will be terminated with a Check Condition status.
The Sense Key must be set to Medium Error and the Additional Sense Code must
be set to Logical Unit Not Ready, Auxiliary Memory Not Accessible (0410h).
If the medium auxiliary memory has failed, the READ ATTRIBUTE command will
be terminated with a Check Condition status. The Sense Key must be set to
Medium Error and the Additional Sense Code must be set to Auxiliary Memory
Read Error (1112h).
The supported Service Action codes are listed in Table 141 on page 157. The
sections that follow the table give the format for each supported service action.

Table 141. Supported Service Action Codes
Code Name Description Format of Returned Data
See “Format for the Attribute
00h Attribute Values Return attribute values Values Service Action” on page
159
See “Format for the Attribute
Returns a list of available
01h Attribute List List Service Action” on page
attribute identifiers
160
See “Format for the Volume
02h Volume List List Service Action” on page
Volume Numbers
161
See “Format for the Partition
03h Partition List List Service Action” on page
Partition Numbers
162
04h Restricted Not applicable Not applicable
05h - 1Fh Reserved Not applicable Not applicable
The First Attribute ID field specifies the attribute identifier of the first attribute to
be returned. If the specified attribute identifier is in the unsupported or
nonexistent state, the READ ATTRIBUTE command will be terminated with a
Check Condition status (see clause 5.10 in the SCSI Primary Commands-3 (SPC-3)).
The Sense Key must be set to Illegal Request and the Additional Sense Code must
be set to Invalid Field in CDB.
The Allocation Length field specifies how many bytes have been allocated for the
returned parameter list. If the length is not sufficient to contain the entire
parameter list, the first portion of the list will be returned. This is not considered
an error. If the remainder of the list is required, the application client should send
a new READ ATTRIBUTE command with an allocation length large enough to
contain the entire parameter list or use the First Attribute ID field to restrict the
attributes that are returned.
The format of parameter data that is returned by the READ ATTRIBUTE command
depends on the service action that is specified.

Vendor-Specific Medium Type Attributes
Table 142 describes the vendor-specific medium type attributes. Application clients
may use Read Attribute to read the contents of the attributes shown in the table.
Table 142. Vendor-Specific Medium Type Attributes
Attribute Identifier Name Attribute Length (in bytes) Format
Unique Cartridge Identity
1000h 28 Binary
(UCI)
The Unique Cartridge Identity (UCI) attribute is only supported on Ultrium 3 and
later drives. The Unique Cartridge Identity (UCI) attribute is read only. Any
attempt to change this attribute by using the Write Attribute command will be
rejected by the drive with a Sense Key 5 ASC/ASCQ 2600 Illegal Request, Invalid
Field In Parameter List. If the cartridge has inconsistency between the various
sources from which the components of the UCI are derived, then any command to
read the UCI will be rejected with a Check Condition, and Sense Key and
ASC/ASCQ of (03/1112), which is an Auxiliary Memory Read Error. This error
shall also be reported if the cartridge has not been initialized.

Format for the Attribute Values Service Action
The READ ATTRIBUTE command with Attribute Values service action returns
parameter data that contains the attributes that are specified by the Partition
Number, Volume Number, and First Attribute ID fields in the CDB. The returned
parameter data contains the requested attributes in ascending numerical order by
attribute identifier value and in the format shown in Table 143.
Table 143. Parameter Data for an Attribute Values Service Action
Bit
Byte 7 6 5 4 3 2 1 0
0
: Available Data (n-3)
3
4
: Attribute #1
x
m
: Attribute #y
The Available Data field will contain the number of bytes of attribute information
in the parameter list. If the parameter list is truncated as a result of insufficient
allocation length, the content of the Available Data field is not altered. The format
of the attribute is described in Table 149 on page 164.

Format for the Attribute List Service Action
The READ ATTRIBUTE command with Attribute List service action returns
parameter data that contains the attribute identifiers for the attributes that are not
in the unsupported state and not in the nonexistent state (for information about
MAM attribute states, see Table 148 on page 163). The contents of First Attribute ID
field in the CDB is ignored. The returned parameter data contains the requested
attribute identifiers in ascending numerical order by attribute identifier value and
in the format shown in Table 144.
Table 144. Parameter Data for an Attribute List Service Action
Bit
Byte 7 6 5 4 3 2 1 0
0
: Available Data (n-3)
3
4
: Attribute ID #1
5
n-1
: Attribute ID #y

Format for the Volume List Service Action
The READ ATTRIBUTE command with Volume List service action returns
parameter data that identifies the supported number of volumes (see Table 145).
The contents of Volume Number, Partition Number, and First Attribute ID fields in
the CDB are ignored.
Table 145. Parameter Data for a Volume List Service Action
Bit
Byte 7 6 5 4 3 2 1 0
0
: Available Data (2)
1
2 First Volume Number (0)
3 Number of Volumes Available (1)
The Available Data field contains two.
The First Volume Number field indicates the first volume that is available and will
be set to 0.
The Number of Volumes Available field indicates the number of volumes that are
available and will be set to 1.

Format for the Partition List Service Action
The READ ATTRIBUTE command with Partition List service action returns
parameter data that identifies the number of partitions that are supported in the
specified volume number (see Table 146). The contents of the Partition Number
and First Attribute ID fields in the CDB are ignored.
Table 146. Parameter Data for a Partition List Service Action
Bit
Byte 7 6 5 4 3 2 1 0
0
: Available Data (2)
1
2 First Partition Number (0)
3 Number of Partitions Available (1)
The Available Data field contains two.
The First Partition Number field indicates the first partition that is available on the
specified volume number and is set to 0.
The Number of Partitions Available field indicates the number of partitions that are
available on the specified volume number and is set to 1.

Medium Auxiliary Memory
Ultrium media includes a non-volatile memory that is referred to as medium
auxiliary memory (MAM). MAM is used to store data that describes the media and
its contents. The Ultrium drives support MAM with the READ ATTRIBUTE and
WRITE ATTRIBUTE commands. The commands retrieve and store information as
attributes in MAM. For more information, see “READ ATTRIBUTE” on page 156
and “WRITE ATTRIBUTE” on page 257.
A MAM attribute is composed of the following components:

v Attribute identifier
v Attribute format code
v Bit that indicates whether the attribute is read only.
v Attribute length that specifies the number of bytes in the attribute’s value
v Value of the attribute
Table 147 lists the three types of MAM attributes.

Table 147. Types of MAM Attributes
Readable with Writable with
Type of MAM
Attribute Source Example READ WRITE
Attribute
ATTRIBUTE ATTRIBUTE
Permanently stored in the MAM Media Serial
Medium Yes No
during manufacture. Number
Device Maintained by the tape drive. Load Count Yes No
Host Maintained by the application client. Backup Date Yes Yes
Table 148 shows the states for the types of MAM attributes.
Table 148. States for the Types of MAM Attributes
Attribute Type Attribute State Description
An application server may read the contents of the attribute with
the READ ATTRIBUTE command, but an attempt to clear or change
the attribute by using the WRITE ATTRIBUTE command will result
Read Only in the command being terminated with a Check Condition status.
Medium or Device When the Read Only field of the MAM attribute is 1, the attribute
is in the read only state. (For information about the Read Only
field, see “Format of MAM Attribute” on page 164.)
The tape drive does not support the attribute and will not return it
Unsupported
in response to a READ ATTRIBUTE command.
A host attribute does not exist in the MAM until a WRITE
Nonexistent
ATTRIBUTE command creates it.
The attribute has been created by using the WRITE ATTRIBUTE
command. After the attribute has been created, the contents may be
Host altered by using subsequent WRITE ATTRIBUTE commands. A
read/write attribute may be returned to the nonexistent state by
Read/Write
using a WRITE ATTRIBUTE command with the Attribute Length
set to 0. When the Read Only field of the MAM attribute is 0, the
attribute is in the read/write state. (For information about the Read
Only field, see “Format of MAM Attribute” on page 164.)

Format of MAM Attribute
Each MAM attribute will be communicated between the application client and tape
drive in the format shown in Table 149. This format will be used in the parameter
data for the WRITE ATTRIBUTE and READ ATTRIBUTE commands. The attribute
format implies nothing about the physical representation of an attribute in the
MAM. For more information, see “READ ATTRIBUTE” on page 156 and “WRITE
ATTRIBUTE” on page 257.
Table 149. Format of a MAM Attribute
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
: Attribute Identifier
1 (LSB)
2 Read Only Reserved Format
3 (MSB)
: Attribute Length (n-4)
4 (LSB)
5 (MSB)
: Attribute Value
n (LSB)
The Attribute Identifier field contains a value that identifies the attribute. For the
range of values in this field, see “Values for the Attribute Identifier Field” on page
165.
The Read Only field indicates whether the attribute is in the read only or
read/write state. If the field is 1, the attribute is in the read only state; if the field
is 0, the attribute is in the read/write state.
The Format field specifies the format of the data in the Attribute Value field.
Table 150 describes the values and requirements for the Format field.
Table 150. Values and Requirements for the Format Field
Generation
Format Name Description
(see Legend)
00b Binary 1,2 The Attribute Value field will contain binary data.
The Attribute Value field will contain only graphic
codes (for example, byte code values 20h through 7Eh),
01b ASCII 1,2 will be left-aligned and place any unused bytes at the
highest offset in the field, and will contain 20h (for
example, ASCII space) in any unused bytes.
The attribute contains textual data. For a description of
10b Text 1,2 the character set, see page 170 and Table 157 on page
170.
11b Reserved N/S The Attribute Value field is reserved.

Table 150. Values and Requirements for the Format Field (continued)
Generation
Format Name Description
(see Legend)
Legend:
1 = supported in IBM TotalStorage LTO Ultrium Tape Drive (commonly called the Ultrium 1 tape drive)
2 = supported in IBM TotalStorage LTO Ultrium 2 Tape Drive (commonly called the Ultrium 2 tape drive)
N/S = not supported
The Attribute Length field specifies the length in bytes of the Attribute Value field.
The value of the Attribute Length field depends on the attribute that is returned.
The Attribute Value field contains the current (READ ATTRIBUTE) or desired
(WRITE ATTRIBUTE) value of the attribute.
Values for the Attribute Identifier Field: The values in the Attribute Identifier
field are assigned according to the attribute type and whether the attribute is
standard or vendor unique. Table 151 lists the range of values for the each attribute
type.
Table 151. Range of Values for MAM Attribute Identifiers
Range of Attribute Identifiers Attribute Type Standard or Vendor Unique
0000h - 03FFh Device Standard
0400h - 07FFh Medium Standard
0800h - 0BFFh Host Standard
0C00h - 0FFFh Device Vendor Unique
1000h - 13FFh Medium Vendor Unique
1400h - 17FFh Host Vendor Unique
1800h - FFFFh Reserved Not applicable
Note: Ultrium drives accept and process a WRITE ATTRIBUTE command that contains standard host type attribute
identifier values (for example 0800h to 0BFFh) or vendor-unique host type attribute identifier values (for example
1400h to 17FFh). Standard host type attribute identifier values may be checked for conformance to the requirements
described in “Standard Host Type Attributes” on page 169. Attributes may be written as long as there is sufficient
available space in MAM. The Ultrium drive supports a total of 1008 bytes of Application Specific Data available for
host use. Each attribute written consumes four bytes of that space for the required attribute header.
The sections that follow describe the standard type attributes for device, medium,
and host.
Standard Device Type Attributes: Device type attributes are maintained and updated
by the drive when the medium and associated MAM are present. Table 152 on
page 166 describes the standard device type attributes.

Table 152. Standard Device Type Attributes
Attribute
Attribute Identifier Name Length Format
(in bytes)
0000h Remaining Capacity in Partition 8 Binary
0001h Maximum Capacity in Partition 8 Binary
0002h Restricted N/A N/A
0003h Load Count 8 Binary
0004h MAM Space Remaining 8 Binary
0005h - 0006h Restricted N/A N/A
0007h Initialization Count 2 Binary
0008h Volume Identifier 32 ASCII
0009h - 0209h Reserved N/A N/A
020Ah Device Make/Serial Number at Last Load 40 ASCII
020Bh Device Make/Serial Number at Load-1 40 ASCII
020Ch Device Make/Serial Number at Load-2 40 ASCII
020Dh Device Make/Serial Number at Load-3 40 ASCII
020Eh - 021Fh Reserved N/A N/A
0220h Total MBytes Written in Medium Life 8 Binary
0221h Total MBytes Read in Medium Life 8 Binary
0222h Total MBytes Written in Current/Last Load 8 Binary
0223h Total MBytes Read in Current/Last Load 8 Binary
0224h - 033Fh Reserved N/A N/A
0340h Medium Usage History (not supported) N/A N/A
0341h Partition Usage History (not supported) N/A N/A
0342h - 03FFh Reserved N/A N/A
Note: N/A = not applicable
Remaining Capacity in Partition and Maximum Capacity in Partition are native

capacities, assuming that there is no data compression for the specified medium
partition. These values are expressed in increments of 1 048 576 bytes (for example,
a value of 1 means 1 048 576 bytes; a value of 2 means 2 097 152 bytes; and so
forth).
Load Count indicates how many times this medium has been fully loaded. This
attribute should not be reset by any action of the device server.
MAM Space Remaining indicates the space that is currently free in the medium
auxiliary memory. The total MAM capacity is reported in the MAM Capacity
attribute. (For a description of the MAM Capacity attribute, see page 169.) Note: It
may not always be possible to use all of the free space that is reported.
Initialization Count indicates the number of times that a device server has logically
formatted the medium. This figure is cumulative over the life of the medium and
will never be reset.

On Ultrium 4 drives, the Volume Identifier indicates the volume identifier of the
medium (set by the library either via the Library Drive Interface or via the ADI
interface). If the volume identifier has not been set, the drive will report this
attribute with an attribute length value of zero.
On Ultrium 1, 2, and 3 drives, the Volume Identifier field is not supported.
The Device Vendor Identification/Product Serial Number at Last Load, Device

Vendor Identification/Product Serial Number at Load-1, Device Vendor
Identification/Product Serial Number at Load-2, and Device Vendor
Identification/Product Serial Number at Load-3 attributes give a rolling history of
the last four device servers in which the medium has been loaded. The format for
the attributes is shown in Table 153 on page 168.

Table 153. Format for Device Vendor Identification/Product Serial Number Attribute, Device Vendor
Identification/Product Serial Number at Load-1 Attribute, Device Vendor Identification/Product Serial Number at
Load-2 Attribute, and Device Vendor Identification/Product Serial Number at Load-3 Attribute
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
: Vendor Identification
7 (LSB)
8 (MSB)
: Product Serial Number
39 (LSB)
The Vendor Identification field will be the same value that is returned in the
Standard Inquiry Data.
The Product Serial Number field contains a vendor-unique serial number. If the
product serial number is not available, the Product Serial Number field will
contain ASCII spaces (20h).
Total MBytes Written in Medium Life and Total MBytes Read in Medium Life
indicate the number of data bytes that are transferred to or from the medium
surface (after any data compression has been applied) over the entire life of the
medium. These values are cumulative and will never be reset. They are expressed
in increments of 1 048 576 bytes (for example, a value of 1 means 1 048 576 bytes;
a value of 2 means 2 097 152 bytes; and so forth).
Total MBytes Written in Current/Last Load and Total MBytes Read in

Current/Last Load indicate the total number of data bytes that are transferred to
or from the medium surface (after any data compression has been applied) during
the current load if the medium is currently loaded, or during the last load if the
medium is currently unloaded. The device server should reset these attributes to 0
when the medium is loaded. These values are expressed in increments of 1 048 576
bytes (for example, a value of 1 means 1 048 576 bytes; a value of 2 means 2 097
152 bytes; and so forth).
Standard Medium Type Attributes: Medium type attributes are hard-coded into the
MAM at the time of manufacture. All supported medium type attributes have a
status of read only. Table 154 describes the standard medium type attributes.
Table 154. Standard Medium Type Attributes
Attribute
(in bytes)
0400h Medium Manufacturer 8 ASCII
0401h Medium Serial Number 32 ASCII
0402h - 0405h Restricted N/A N/A
0406h Medium Manufacture Date 8 ASCII
0407h MAM Capacity 8 Binary
0408h Medium Type 1 Binary

Table 154. Standard Medium Type Attributes (continued)
Attribute
(in bytes)
0409h Medium Type Information 2 Binary
Numeric Medium Serial Number (not
040Ah N/A N/A
supported)
040Bh - 07FFh Reserved N/A N/A
Medium Manufacturer contains 8 bytes of ASCII data that identifies the vendor of
the media.
Medium Serial Number identifies the manufacturer’s serial number for the
medium.
Medium Manufacture Date identifies the date of manufacture of the medium. The
format is YYYYMMDD (four numeric ASCII characters for the year, followed by two
numeric ASCII characters for the month, followed by two numeric ASCII
characters for the day, with no intervening spaces).
MAM Capacity is the total capacity of the medium auxiliary memory (in bytes) at
the time of manufacture. It does not indicate the free space of unused MAM
because some of the MAM space may be reserved for device-specific use, which
makes it inaccessible to the application client.
Medium Type and Medium Type Information give information about non-data
media and other types of media. The Medium Type Information attribute is
interpreted according to the type of medium that is indicated by the Medium Type
attribute. Table 155 give the values for the Medium Type and Medium Type
Information attributes.
Table 155. Values for Medium Type and Medium Type Information Attributes
Medium Type Description Medium Type Information
00h Data medium Reserved
01h Cleaning medium Maximum number of cleaning cycles permitted
02h - 7Fh Reserved Reserved
80h Write-once medium Reserved
81h - FFh Reserved Reserved
Standard Host Type Attributes: Table 156 describes the standard host type attributes.
Application clients may use the WRITE ATTRIBUTE and READ ATTRIBUTE
commands to maintain the attributes shown in the table. All existing host type
attributes have a status of read/write.
Table 156. Standard Host Type Attributes
Attribute
(in bytes)
0800h Application Vendor 8 ASCII

Table 156. Standard Host Type Attributes (continued)
Attribute
(in bytes)
0801h Application Name 32 ASCII
0802h Application Version 8 ASCII
0803h User Medium Text Label 160 Text
0804h Date and Time Last Written 12 ASCII
0805h Text Localization Identifier 1 Binary
0806h Barcode 32 ASCII
0807h Owning Host Textual Name 80 Text
0808h Media Pool 160 Text
0809h Partition User Text Label (not supported) N/A N/A
080Ah Load/Unload at Partition (not supported) 1 Binary
080Bh - BFFh Reserved N/A N/A
Application Vendor contains 8 bytes of ASCII data that identifies the manufacturer
of the application client (for example, a class driver or backup program) that most
recently sent a WRITE ATTRIBUTE command to the tape drive while this MAM
was accessible.
Application Name contains the name of the application client.
Application Version contains the version of the application client.
User Medium Text Label is the user-level identifier for the medium.
Date and Time Last Written contains when the application client last wrote to the
MAM. The format is YYYYMMDDHHMM (four numeric ASCII characters for the year,
followed by two numeric ASCII characters for the month, followed by two
numeric ASCII characters for the day, followed by two numeric ASCII characters
between 00 and 24 for the hour, followed by two numeric ASCII characters for the
minute, with no intervening spaces).
Text Localization Identifier defines the character set that is used for attributes with
a Text format. Table 157 gives the values for the Text Localization Identifier
attribute.
Table 157. Values for the Text Localization Identifier Attribute
Value Meaning
00h No code specified (ASCII)
01h ISO/IEC 8859-1 (Europe, Latin America)
02h ISO/IEC 8859-2 (Eastern Europe)
03h ISO/IEC 8859-3 (Southeastern Europe, miscellaneous)
04h ISO/IEC 8859-4 (Scandinavia/Baltic)
05h ISO/IEC 8859-5 (Cyrillic)
06h ISO/IEC 8859-6 (Arabic)

Table 157. Values for the Text Localization Identifier Attribute (continued)
Value Meaning
07h ISO/IEC 8859-7 (Greek)
08h ISO/IEC 8859-8 (Hebrew)
09h ISO/IEC 8859-9 (Latin 5)
0Ah ISO/IEC 8859-10 (Latin 6)
0Bh - 7Fh Reserved
80h ISO/IEC 10646-1 (UCS-2BE)
81h ISO/IEC 10646-1 (UTF-8)
82h - FFh Reserved
Barcode is the contents of a bar code that is associated with the medium in the
MAM.
Owning Host Textual Name indicates the host from which the User Medium Text
label originates.
Media Pool indicates the media pool to which this medium belongs.
Vendor-Specific Medium Type Attributes: Table 158 describes the vendor-specific

medium type attributes. Application clients may use Read Attribute to read the
contents of the attributes shown in the table.
Table 158. Vendor-Specific Medium Type Attributes
Attribute
Length
Attribute Identifier Name (in bytes) Format
1000h Unique Cartridge Identity (UCI) 28 Binary
The Unique Cartridge Identity (UCI) attribute is supported only on Ultrium 3

drives. The Unique Cartridge Identity (ICU) attribute is read only. Any attempt to
access it using the Write Attribute command will be rejected by the drive with a
Sense Key 5 ASC/ASCQ 2600 Illegal Request, Invalid Field In Parameter List. If
the cartridge has inconsistency between the various sources from which the
components of the UCI are derived, then any command to read the UCI will be
rejected with a Check Condition, and Sense Key and ASC/ASCQ of (03/1112),
which is an Auxiliary Memory Read Error. This error shall also be reported if the
cartridge has not been initialized.

READ BLOCK LIMITS
The READ BLOCK LIMITS command (see Table 159) requests that the READ
BLOCK LIMITS data (see Table 160) be returned. The READ BLOCK LIMITS data
specifies the drive’s limit on block lengths.
Table 159. READ BLOCK LIMITS Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved
2 Reserved
3 Reserved
4 Reserved
5 Control
The format of the data returned in the READ BLOCK LIMITS Descriptor is shown
in Table 160.
Table 160. READ BLOCK LIMITS Descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0 Reserved Granularity (0h)
1
: Maximum Block Length Limit (FFFFFFh)
3
4
: Minimum Block Length Limit (0001h)
The Granularity field specifies the supported block size granularity. For Ultrium
drives this is set to 0, which indicates that the drive supports all block sizes equal
to n, where n is greater than or equal to the Minimum Block Length Limit and less
than or equal to the Maximum Block Length Limit.
The Maximum Block Length Limit is set to 0xFFFFFF.
The Minimum Block Length Limit is set to 1.
For READ and WRITE commands with the Fixed field set to 1, block lengths are
limited to multiples of four.
The Ultrium drives support fixed-block transfers or variable-block transfers, with

the block length constrained between the given limits in either transfer mode. The
transfer mode is controlled by the Fixed field in the WRITE or READ commands.

READ BUFFER
The READ BUFFER command reads data from the memory on the drive and sends
it to the initiator. The command is used in conjunction with the WRITE BUFFER
command as a diagnostic function for testing memory in the drive and the
integrity of the service delivery subsystem. The READ BUFFER command is also
used for retrieving data that is specified by the value of the Buffer ID. This
command does not alter the medium.
Table 161. READ BUFFER Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Mode
2 Buffer ID
3
: Buffer Offset
5
6
: Allocation Length
8
9 Control
The Mode field and its meaning are described in Table 162.
Table 162. Description of the Mode Field
Mode Description Support
00h Combined header and data 1, 2, 3, 4
02h Data 1, 2, 3, 4
03h Descriptor 1, 2, 3, 4
07h Descriptor (see Note) 1, 2, 3, 4
0Ah Echo buffer 2, 3, 4
0Bh Echo buffer descriptor 2, 3, 4
Legend:
Note: The descriptor that is returned for Mode 07h has the Buffer Capacity field reported in 64-byte increments.
The Buffer ID indicates which buffer is to be read. The Buffer IDs are shown in
Table 166 on page 176.
The Buffer Offset field may be set to any address in the buffer.

If the Mode is set to 07h and the Buffer ID is 0, the descriptor that is returned is
for the Main Data buffer, and the Buffer Capacity field is the number of 64-byte
segments that are available.
The format of the 4-byte descriptor is shown in Table 164.
Note: The Main Data buffer capacity is larger than can be represented in the Buffer
Offset field of the CDB and Buffer Capacity field of the header. To
compensate for this, the Buffer Offset and Buffer Capacity fields for the
Main Buffer (buffer ID = 0) are interpreted and expressed in multiples of 64
bytes (for example, a value of 1 equals 64 bytes). This interpretation is for
buffer modes 00h, 01h, 02h, 03h, and 07h only.
If the Mode is set to 0Ah, data from the echo buffer is returned.
In this mode, Buffer ID and Buffer Offset fields are ignored. Prior to issuing a
READ BUFFER command that uses the echo buffer, a WRITE BUFFER command
that uses the echo buffer must have been successfully completed (see“WRITE
BUFFER” on page 258). If not, the Read Echo Buffer terminates with a Check
Condition status, the Sense Key is set to Illegal Request, and ASC/ASCQ is set to
Command Sequence Error (2C00h). The Read Echo Buffer returns the same number
of bytes of data as was received in the prior Write Echo Buffer from the same
initiator.
If the Mode is set to 0Bh, the descriptor information of the echo buffer is returned.
The format of the echo buffer descriptor is shown in Table 165 on page 175. The
Echo Buffer Overwritten Supported (EBOS) is set to 1 because the drive keeps the
echo buffer for each initiator.
Table 163. READ BUFFER Header
Bit
Byte 7 6 5 4 3 2 1 0
0 Reserved
1
: Buffer Capacity
Table 164. READ BUFFER Descriptor

Bit
Byte 7 6 5 4 3 2 1 0
0 Offset Boundary (0h means byte boundary)
1
: Buffer Capacity

Table 165. READ ECHO BUFFER Descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0 Reserved EBOS
(1)
1 Reserved
2 Reserved Buffer Capacity
3 Buffer Capacity

Table 166. Drive Buffers
Supported Buffers ID Offset Boundary Ultrium Support
Main Data 00h 4 1, 2, 3, 4
Dump Data (Read Only) (See Note) 01h 4 1, 2, 3, 4
Test 02h 4 1, 2, 3, 4
VPD 03h 4 1, 2, 3, 4
Firmware 04h 4 1, 2, 3, 4
Cartridge Memory (Read Only) 05h 4 1, 2, 3, 4
Error Log (Read Only) 06h 4 1, 2, 3, 4
SCSI Log (Read Only) 07h 4 1, 2, 3, 4
08h 4 1, 2, 3, 4
Reserved 09h-80h N/A N/A
(IBM use only) 81h N/A N/A
(IBM use only) 83h N/A N/A
Reserved 84h-FFh N/A N/A
Legend:
N/A = not applicable
Note: When reading dump data on a Fibre Channel drive, take care to ensure that the amount of data in each
transfer does not exceed the DMA settings of the HBA. The dump data currently exceeds 1 MB in size. The
default DMA setting for some HBAs is 1 MB. For this reason, it is recommended that the dump data is read in a
series of smaller blocks with appropriate offsets (for example, 64 K bytes).
Note: When reading dump data the data should be read with strictly increasing offsets. If it is not read with
strictly increasing offsets and a new dump is created during the read processing, the data might be corrupted

Error Log Buffer (06h)
The error log buffer contains zero or more entries described in Table 167.
Table 167. Error Log Buffer Command
Bit
Byte 7 6 5 4 3 2 1 0
0
: Time Stamp
3
4 Entry Number
5 Error Code
6
: Fsc 1st Text
7
8
: Fsc 1st Data
9
10
: Fsc 2nd Text
11
12
: Fsc 2nd Data
13
14
: Cartridge Serial Number
21
22
: EC Level
27
28
: Hardware Level
31

World Wide Name Buffer (08h)
This buffer contains the World Wide Name values that are used by the drive on
% LUN 0. Table 168 describes the World Wide Name buffer.
Table 168. World Wide Name Buffer
Bit
Byte 7 6 5 4 3 2 1 0
0
: World Wide Node Name
7
8
: World Wide Port Name
15

SCSI Log Buffer (07h)
The Error Log Buffer contains 10 entries, each of which has the format described in
Table 169.
Table 169. Error Log Buffer Command
Bit
Byte 7 6 5 4 3 2 1 0
0
: Time Stamp (seconds)
3
4 Entry Number
5 Initiator SCSI ID
6 SCSI Status
7 Reserved
8
: CDB
23
24
: Sense Data
59
An Entry Number set to 0 indicates that the entry is not valid. The scheme used
for setting the Entry Number is not described in this manual.
Valid entries are built for commands that are issued to LUN 0 and that get a Check
Condition status for sense data that contain a Sense Key of 3 or 4.
The CDB field contains the contents of the CDB that received Check Condition
status even when the check condition is a Deferred Check Condition.

READ POSITION
The READ POSITION command returns current position information to the
initiator. It can be used to find the current logical position of the medium and to
find information about the number of bytes or blocks in the buffer. Table 171
shows the format of the returned data.
Table 170. READ POSITION Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Service Action (0)
2 Reserved
3 Reserved
4 Reserved
5 Reserved
6 Reserved
7
: Parameter Length (0)
8
9 Control

Table 171. READ POSITION Data
Bit
Byte 7 6 5 4 3 2 1 0
0 BOP EOP BCU BYCU(1) Rsvd BPU(0) PERR(0) Rsvd
2 Reserved
3 Reserved
4
: First Block Location
7
8
: Last Block Location
11
12 Reserved
13
: Number of blocks in buffer
15
16
: Number of bytes in buffer (0)
19
The First Block Location specifies the block address that is associated with the
current logical position. The value indicates the block address of the next data
block to be transferred between the initiator and the target if a READ or WRITE
command is issued.
The Last Block Location is specified by the following procedure:

1. After a WRITE, WRITE FILEMARK, or any command failed for a deferred
write error, this field specifies the block address that is associated with the next
block to be transferred from the buffer to the medium.
2. After a successful read type command (for example Read, Space, Locate,
non-immediate Rewind), this field is returned as 0.
The Beginning of Partition (BOP) field is set if the Block Location fields are 0.
An end-of-partition (EOP) field of 1 specifies that the logical unit is positioned

between early-warning and end-of-partition. An EOP field of 0 specifies that the
current logical position is not between early-warning.
The Block Position Unknown (BPU) field is always set to 0, because the position is
always known.
The Partition Number is set to 0 because partitioning is not supported.

A block count unknown (BCU) field of 1 indicates that the Number of Blocks in
Buffer field does not represent the actual number of blocks in the buffer. A BCU
field of 0 indicates that the Number of Blocks in Buffer field is valid.
A byte count unknown (BYCU) field of 1 indicates that the Number of Bytes in
Buffer field does not represent the actual number of bytes in the buffer. This field
is always set to 1.
The Number of blocks in buffer field is specified by the following procedure:

1. After WRITE, WRITE FILEMARK, or any command failed for a deferred write
error, this field specifies the number of unwritten data blocks and filemarks
that are still in the buffer.
2. After successful read type commands this field will always return 0.

RECEIVE DIAGNOSTIC RESULTS
The RECEIVE DIAGNOSTIC RESULTS command returns the results of diagnostic
tests to the initiator. The format of the data returned is specified in “SEND
DIAGNOSTIC” on page 223.
Table 172. RECEIVE DIAGNOSTIC RESULTS Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved
2 Reserved
3
: Allocation Length
4
5 Control

RELEASE UNIT
The RELEASE UNIT command removes a reservation made by a RESERVE UNIT
command. If there is an existing reservation from the same initiator with the same
parameters, then that reservation is removed and Good status is returned. It is not
an error to attempt to release a reservation that is not currently valid or is held by
another initiator. In this case, the drive returns Good status without altering any
reservation.
Table 173. 6-Byte RELEASE UNIT Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Obsolete
2 Obsolete
3 Reserved
4 Obsolete (0)
5 Control
Table 174. 10-Byte RELEASE UNIT Command

Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete 3rdPty (0) LongID (0) Reserved
2 Reserved
3 Third Party Device ID (0)
4 Reserved
5 Reserved
6 Reserved
7
: Parameter List Length (0)
8
9 Control

REPORT DENSITY SUPPORT
The REPORT DENSITY SUPPORT command returns details about the tape formats
supported by the drive. The data is returned as a header and a series of descriptor
blocks. If the Media field is set, then one descriptor block is returned with the data
for the currently loaded tape. If the Media field is set to 0, the density support data
block descriptors are returned by ascending Primary Density Code values.
Table 175. REPORT DENSITY SUPPORT Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved Media
2 Reserved
3 Reserved
4 Reserved
5 Reserved
6 Reserved
7
: Allocation Length
8
9 Control
The Allocation Length field specifies the maximum number of bytes that the device
server may return.
The REPORT DENSITY SUPPORT command returns the REPORT DENSITY

SUPPORT header (see Table 176 on page 186) followed by one or more density
support data block descriptors (see Table 177 on page 186). The density support
data block descriptors follow the density support header.
In an Ultrium 1 drive, the Ultrium 1 descriptor is always returned with the DEFLT
field set to 1.
In an Ultrium 2 drive, if a medium is loaded in the drive and the Media field is set
to 1, the descriptor of the loaded medium will be returned with the DEFLT field
set to 1. If the Media bit is set to 0, the density support data block descriptors are
returned with the Ultrium 1 descriptor followed by the Ultrium 2 descriptor. If
there is no medium in the drive and the Media field is set to 0, both descriptors
will have the DEFLT field set to 1. If a medium is loaded in the drive and the
Media field is set to 0, the descriptor of the density of the loaded medium will be
returned with the DEFLT field set to 1 and the descriptor of other density will be
returned with the DEFLT bit set to 0.
returned with the Ultrium 1 descriptor, followed by the Ultrium 2 descriptor,
followed by the Ultrium 3 descriptor. If there is no medium in the drive and the
Media field is set to 0, all descriptors will have the DEFLT field set to 1. If a

medium is loaded in the drive and the Media field is set to 0, the descriptor of the
density of the loaded medium will be returned with the DEFLT field set to 1 and
the descriptor of other density will be returned with the DEFLT bit set to 0.
The format of the REPORT DENSITY SUPPORT header is as follows:
returned with the Ultrium 2 descriptor, followed by the Ultrium 3 descriptor,
followed by the Ultrium 4 descriptor. If there is no medium in the drive and the
Media field is set to 0, all descriptors will have the DEFLT field set to 1. If a
medium is loaded in the drive and the Media field is set to 0, the descriptor of the
density of the loaded medium will be returned with the DEFLT field set to 1 and
the descriptor of other density will be returned with the DEFLT bit set to 0.
Table 176. REPORT DENSITY SUPPORT Header
Bit
Byte 7 6 5 4 3 2 1 0
0
: Available Density Descriptor Length
1
2 Reserved
3 Reserved
The Available Density Descriptor Length gives the total amount of data that is
available to be returned and does not include itself.
The header is followed by one or more REPORT DENSITY SUPPORT descriptor

blocks with the format in Table 177.
Table 177. REPORT DENSITY SUPPORT Descriptor Block
Bit
Byte 7 6 5 4 3 2 1 0
0 Primary Density Code
1 Secondary Density Code
2 WRTOK DUP(0) DEFLT Reserved
3 Reserved
4 Reserved
5
: Bits per mm
7
8
: Media Width

Table 177. REPORT DENSITY SUPPORT Descriptor Block (continued)
Bit
Byte 7 6 5 4 3 2 1 0
10
: Tracks
11
12
: Capacity
15
16
: Assigning Organization
23
24
: Density Name
31
32
: Description
51
Table 178 shows the values that are reported for the Ultrium format.
Table 178. Density Information for LTO Formats
Field Ultrium 1 Ultrium 2 Ultrium 3 Ultrium 4
Primary density code 40h 42h 44h 46h
Secondary density code 40h 42h 44h 46h
Bits per mm 4880 7398 9638 12725
Media width (in tenths of mm) 127 127 127 127
Tracks 384 512 704 896
95,367 190,734 381,469 800,000
Capacity1
(in 220 bytes) (in 220 bytes) (in 220 bytes) (in 106 bytes)
Assigning organization "LTO-CVE " "LTO-CVE " "LTO-CVE " "LTO-CVE "
Density name "U-18 " "U-28 " "U-316" "U-416 "
Description "Ultrium 1/8T" "Ultrium 2/8T" "Ultrium 3/16T" "Ultrium 4/16T "
Notes:
1: See each generation value to see what units the capacity is expressed in. Earlier generations express
Capacity in non-standard units.
The Write Okay (WRTOK) field is set to 0 if the drive does not support writing to
this format, but does support reading it. This is always set to 1 for Ultrium 1 and
Ultrium 2 drives. In Ultrium 3 drives, the WRTOK field is set to 0 when
Generation 1 media is in the drive, and set to 1 when Generation 2 or Generation 3

media is in the drive. In Ultrium 4 drives, the WRTOK field is set to 0 when
Generation 1 or Generation 2 media is in the drive, and set to 1 when Generation 3
or Generation 4 media is in the drive.
The Duplicate (DUP) field is set to 0 for every descriptor block, indicating that
each density is reported only once. A DEFLT field of 0 specifies that this density is
not the default density of the drive. A DEFLT field of 1 specifies that this density is
the default density.
Note: The default density of the drive will vary, depending on the currently
mounted media. Multiple codes may return a DEFLT field of 1 when the
Media field is 0 because more than one default is possible.
If the Media field is set to 0, the maximum values possible are reported. In Ultrium
2 drives, the Ultrium 1 descriptor is returned, followed by the Ultrium 2
descriptor. In Ultrium 3 drives, the Ultrium 1 descriptor is returned, followed by
the Ultrium 2 descriptor, followed by the Ultrium 3 descriptor.In Ultrium 4 drives,
the Ultrium 1 descriptor is returned, followed by the Ultrium 2 descriptor,
followed by the Ultrium 3 descriptor, followed by the Ultrium 4 descriptor.
If the Media field is set to 1, the Capacity field specifies the approximate capacity
of the current tape, assuming that recording occurs in this density with one
partition.
If the Media field is 1 and the logical unit is not in the ready state, Check
Condition status will be returned. The Sense Key must be set to Not Ready and
the Additional Sense Code will specify the reason for Not Ready.
The Bits per mm field specifies the number of bits per millimeter per track as
recorded on the medium. See Table 178 on page 187 for the values that are
returned.
The Media Width field specifies the width of the medium that is supported by this
density. See Table 178 on page 187 for the values that are returned.
The Tracks field specifies the number of data tracks that are supported on the
medium by this density. See Table 178 on page 187 for the values that are returned.
If the Media field is 0, the Capacity field specifies the approximate capacity of the
longest supported medium for this density. If the Media field is 1, the Capacity
field specifies the approximate capacity of the current medium for this density. If
the approximate capacity of the current medium is not available for the mounted
medium, the longest supported medium capacity for this density is used. The
capacity assumes that compression is disabled. The capacity also assumes that the
media is in good condition, and that normal data and block sizes are used. This
value is in units of megabytes (106 bytes). The drive does not guarantee that this
space is actually available in all cases. See Table 178 on page 187 for the values that
are returned.
The Assigning Organization field contains 8 bytes of ASCII data that identifies the
organization that is responsible for the specifications that define the values in this
density support data block descriptor. The data is left-aligned within this field. The
ASCII value for a space (20h) is used for padding. See Table 178 on page 187 for
the values that are returned.

The Density Name field contains 8 bytes of ASCII data that identifies the
document (or other identifying name) that is associated with this density support
data block descriptor. The data is left-aligned within this field. The ASCII value for
a space (20h) is used for padding. See Table 178 on page 187 for the values that are
returned.
The Description field contains 20 bytes of ASCII data that describe the density. The
data is left-aligned within this field. The ASCII value for a space (20h) is used for
padding. See Table 178 on page 187 for the values that are returned.

REPORT LUNs
The server uses the REPORT LUNs command to retrieve information about the
Logical Units that the drive supports.
Table 179. REPORT LUNs Command
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation Code (A0h)
1 Reserved
2 Reserved
3 Reserved
4 Reserved
5 Reserved
6
: Allocation Length
9
10 Reserved
11 Control
The allocation length is at least 16 bytes. If this is not the case, the drive returns
Check Condition status, with a Sense Key of Illegal Request and an ASC/ASCQ of
Invalid Field in CDB.
Table 180 shows the data that is returned:

Table 180. Logical Unit Numbers Data
Bit
Byte 7 6 5 4 3 2 1 0
0
: LUN List Length
3
4
: Reserved
7
8
: First LUN (0000000000000000h)
15
n
: Additional LUN descriptor(s)

The LUN List Length field contains the length in bytes of the LUN list that is
available to be transferred. The LUN list length is the number of logical unit
numbers in the logical unit inventory multiplied by eight. This value depends on
drive configuration.
On drives not using the ADI interface that do not have the library control path
feature enabled, this value is 8. On drives not using the ADI interface that have the
library control path feature enabled, this value is 16 and an Additional LUN
descriptor is returned with a value of 0001000000000000h.
On drives using the ADI interface, if the ENABLE bit of Logical Unit Index 01h in
the Mode Page 0Eh subpage 03h is set, then an Additional LUN descriptor is
returned with a value of 0001000000000000h.
On drives using the ADI interface:

v If this command is received over a primary port, and if the ENABLE bit of
Logical Unit Index 02h in Mode Page 0Eh subpage 03h is set, then an Additional
LUN descriptor is returned with a value of 0002000000000000h.
v If this command is received over the ADT port , then an Additional LUN
descriptor is returned with a value of 0002000000000000h, regardless of the
setting of the ENABLE bit of Logical Unit Index 02h in Mode Page 0Eh subpage
03h.
If the allocation length in the CDB is too small to transfer information about the
entire logical unit inventory, the LUN list length value will not be adjusted to
reflect the truncation.

REPORT SUPPORTED TASK MANAGEMENT FUNCTIONS (Not LTO1)
The REPORT SUPPORTED TASK MANAGEMENT FUNCTIONS command (see
Table 181) requests information on task management functions supported by the
drive. This command is not supported on U160 drives.
Table 181. REPORT SUPPORTED TASK MANAGEMENT FUNCTIONS command
7 6 5 4 3 2 1 0
0 Operation Code (A3h)
1 Reserved Service Action (0Dh)
2
Reserved
5
6 (MSB)
Allocated Length (4h or larger)
9 (LSB)
10 Reserved
11 Control
The ALLOCATION LENGTH field indicates how many bytes have been allocated
for the returned parameter data. If the length is not sufficient to contain all the
parameter data, the first portion of the data is returned. This is not considered an
error. The actual length of the parameter data is available in the TIMESTAMP
PARAMETER DATA LENGTH field in the parameter data. If the remainder of the
parameter data is required, the application client should send a new REPORT
TIMESTAMP command with an ALLOCATION LENGTH field large enough to
contain all the data.
The format of the parameter data returned by the REPORT TASK MANAGEMENT
FUNCTIONS command is shown in Table 182.
Table 182. REPORT SUPPORTED TASK MANAGEMENT FUNCTIONS parameter data
7 6 5 4 3 2 1 0
0 ATS ATSS COCAS CTS LURKS ITS TRS WAKES
1 Reserved INTERS
2
Reserved
3

REPORT TIMESTAMP (Not LOT 1)
The REPORT TIMESTAMP command see (see Table 183) requests that the drive
return the value of the logical unit’s timestamp.
Table 183. REPORT TIMESTAMP command
Bit
Byte 7 6 5 4 3 2 1 0
0 OPERATION CODE aha
1 Reserved SERVICE ACTION FHA
2
Reserved
5
6
ALLOCATION LENGTH
9
10 Reserved
11 CONTROL
The ALLOCATION LENGTH field indicates how many bytes have been allocated
for the returned parameter data. If the length is not sufficient to contain all the
parameter data, the first portion of the data is returned. This is not considered an
error. The actual length of the parameter data is available in the TIMESTAMP
PARAMETER DATA LENGTH field in the parameter data. If the remainder of the
parameter data is required, the application client should send a new REPORT
TIMESTAMP command with an ALLOCATION LENGTH field large enough to
contain all the data.
The format of the parameter data returned by the REPORT TIMESTAMP command
is shown in Table 184.
Table 184. Timestamp Descriptor
Bit
Byte 7 6 5 4 3 2 1 0
0
TIMESTAMP PARAMETER DATA LENGTH aha
1
2 Reserved Timestamp Origin
3 Reserved
4 (MSB)
TIMESTAMP
9 (LSB)
10 Reserved
11 Reserved
The TIMESTAMP PARAMETER DATA LENGTH field indicates the number of

bytes of parameter data that follow.
The timestamp origin field indicates the origin of the timestamp returned (see
“Device Clocks” on page 265). The TIMESTAMP field shall contain the current
value of the timestamp (see “Device Clocks” on page 265).

REQUEST SENSE
Table 185. REQUEST SENSE Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved
2 Reserved
3 Reserved
4 Allocation Length
5 Control
The sense data returned is described in Table 186.
Sense Data Format

Table 186. Sense Data Format
Bit
Byte 7 6 5 4 3 2 1 0
0 Valid Sense Error Code
1 Segment Number (0)
2 File Mark EOM ILI Reserved Sense Key
3
: Information
6
7 Additional Sense Length
8
: Command Specific Information
11
12 Additional Sense Code
13 Additional Sense Code Qualifier
14 Field Replaceable Unit Code
15 SKSV C/D Reserved BPV Bit Pointer
16
SKSV (1: Field Pointer)
:
SKSV (0: Reporting Error Fault Symptom Code)
17
18
: Reporting Error Flag Data
19
20 Reserved (0)
21 LowPerf DrvSrvc Reserved

Table 186. Sense Data Format (continued)
Bit
Byte 7 6 5 4 3 2 1 0
22
: Volume Label
28
29 Physical Wrap
30
: Relative LPOS Value
33
34 SCSI Address
35 RS422 Information
Port Identifier (Relative Target Port Address) Reporting Sense (This is the port address of the drive port
through which sense is being reported.
36 On Fibre Channel drives, it is the Fibre Channel Fabric Port Address [for example, 011E13 or 000026]
with byte 36 being reserved.
:
On SAS drives, it is the Hashed SAS Address of the drive port [for example, F32A94] with byte 36 being
39 reserved.
On SCSI, bytes 36 through 38 are reserved, and byte 39 is set to the port’s SCSI address [i.e., byte 39 =
byte 34].)
40 Relative Tgt Port Reporting Sense
Tape 0: Reserved
Directory Reserved Reserved Reserved Reserved
1: Relative Tgt Port 1 (Port 0)
Valid 2: Relative Tgt Port 2 (Port 1)
3: Relative Tgt Port 3 (Library Port)
41 Host Command (SCSI Opcode)
42 Density Type
0: No media present
1: Gen1 (384 track) Media Type (Vendor Reserved)
2: Gen2 (512 track)
3: Gen3 (704 track)
4: Gen 4
43
Volume Label Cartridge Type
44
45
Logical Block Number
:
(Current LBA that would be reported in Read Position command)
48
49
: Data set Number
52
53
1st Error FSC
54

Table 186. Sense Data Format (continued)
Bit
Byte 7 6 5 4 3 2 1 0
55
1st Error Flag Data
56
57
2nd Error FSC
58
59
2nd Error Flag Data
60
61
Next-to-Last Error FSC
62
63
Next-to-Last Error Flag Data
64
65
Last Error FSC
66
67
Last Error Flag Data
68
69 LPOS Region
70
: ERP Summary Information
85
86
Cartridge Serial Number
:
(This is not the Barcode value, but the value from the CM right justified)
95
90
: Reserved (0)
95
The Valid field is set if the Information field contains valid information.
The descriptions that follow serve only as an overview of sense reporting in the
tape drive. This tape drive conforms to all sense field reporting, as specified in the
SCSI Primary Commands-2 (SPC-2).
The Error Code field is set to 70h to indicate a current error that is associated with
the most recently received command. It is set to 71h to indicate a deferred error
that is not associated with the current command.
The segment number is 0, because the COPY, COMPARE, and COPY and VERIFY
commands are not supported.
The File Mark field is set if a SPACE, READ, or VERIFY command did not
complete because a file mark was read.

The End of Medium (EOM) field is set if a WRITE or WRITE FILE MARKS
command completed in the early warning area. Spacing into BOM also causes this
field to be set. It is also set on an attempt to read or space past EOD or if an
attempt is made to space into Beginning of Media.
The Illegal Length Indicator (ILI) field is set if a READ or VERIFY ended because a
block was read from tape that did not have the block length requested in the
command.
For values of the Sense Key field see Chapter 7, “Sense Keys and Additional
Sense,” on page 269.
The Information Bytes are only valid if the Valid field is set. This occurs only for
current errors and not for deferred errors. See the specific command for details
about when Information Bytes are valid.
The Additional Sense Length is set to n-7, and is at least 10. When the sense data is
associated with an Illegal Length read, the Additional Sense Length may be 10. In
Generation 1 and Generation 2 drives, n can be as large as 35. In Generation 3
drives, n can be as large as 95. While this length in Generation 3 drives is not
anticipated to change, it is recommended that the Additional Sense Length be used
to parse that data.
The Command Specific Information is set to 0, because no supported commands

define a use for this field.
For supported Additional Sense Codes and Additional Sense Code Qualifiers, see
Chapter 7, “Sense Keys and Additional Sense,” on page 269.
The Field Replaceable Unit field is set to 0 or to a non-zero, vendor-specific code

that indicates the part of the drive that is suspected of causing the failure.
The only Sense Key-specific data supported is for Illegal Request (5h). For this
sense key, the Sense Key Specific Valid field is set and the following fields may be
set:
v The Command/Data (C/D) field is set to 1 if the illegal parameter was detected
in the Command Descriptor Block, and is set to 0 if it was detected in the Data
phase.
v If a bit within a byte was invalid, the Bit Pointer Valid (BPV) field is set and the
Bit Pointer field is set to indicate which bit was in error.
v The Field Pointer is set to indicate which byte was in error.
The Drive Service (DrvSrvc) bit is mapped to the “Dead Drive” flag of “Log Page
3Ch: Drive Usage Information” on page “Log Page 3Ch: Drive Usage Information”
on page 100. This bit is set to one by code whenever the drive determines that it
has a hardware fault causing the drive to be inoperative.
1. This bit is maintained across all reset conditions, firmware downloads, and
power cycles.
2. This bit does not affect drive operations.
3. The drive Power-On-Test does not test or affect this bit.
The Clean (CLN) field is set if the drive needs cleaning, and is otherwise set to 0.

In a Low Performance drive, the LowPerf bit is set to 1. In a normal,
non-performance limited drive, the LowPerf bit is set to 0.
The Dump field indicates that the drive has a Dump available. The field is used to
indicate when it is appropriate to read a dump.
The Volume Label Fields Valid (VolValid) field is set if the Volume Label being
reported is valid.
If a cartridge is loaded in the drive and the Volume Label Fields Valid is set, the
Volume Label field reports the seven characters from the left of the volume label
from the CM Mechanism Related Data page (if one exists), or it reports the seven
characters from the left of the volume label from the host bar code field in the CM
(if it exists), or it reports all spaces (ASCII 20h).
The Current Wrap reports the physical wrap of the tape. The least significant bit
reflects the current physical direction. A 0 means the current direction is away from
physical beginning of tape. A 1 means the current direction is towards physical
beginning of tape.
Relative LPOS reports the current physical position on tape.
SCSI Address reports the SCSI Bus Address for the drive. Values returned range
from 00h to 0Fh.
The RS422 Information field may contain a value passed across the RS-422 serial
interface by, for example, a tape library, if the library vendor chooses to send such
a value. The value passed from across the RS-422 interface is reported persistently
until a different value is sent, at which time the new value is reported persistently.
The Volume Label Cartridge Type is defined by Table 187.

Table 187. Volume Label Cartridge Type
Definition Native Capacity Volume Label Cartridge Type
Ultrium 1 Type A 100 GB L1
Ultrium 1 Type B 50 GB LA
Ultrium 1 Type C 30 GB LB
Ultrium 1 Type D 10 GB LC
Ultrium 3 WORM A 400 GB WORM LT
Ultrium 4 WORM A 800 GB WORM LU

RESERVE UNIT
The RESERVE UNIT command creates a reservation for the drive. Third-party
reserves are not supported.
Table 188. 6-Byte RESERVE UNIT Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Obsolete (00h)
2 Reserved
3 Reserved
4 Reserved
5 Control
Table 189. 10-Byte RESERVE UNIT Command

Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete 3rdPty(0) Reserved LongID (0) Reserved
2 Reserved
3 Third Party Device ID (0)
4 Reserved
5 Reserved
6 Reserved
7
: Parameter List Length (0)
8
9 Control

REWIND
The REWIND command causes the logical position to be set to BOM.
Table 190. REWIND Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved Immed
2 Reserved
3 Reserved
4 Reserved
5 Control
completed.
On eServer enabled drives (i.e., The Product ID returned in Standard Inquiry is

"HH LTO Gen x" where "x" is a number), if certain criteria are met, a Check
Condition, Sense Key 1h, ASC/ASCQ 8252h (Degraded Media) will be returned
after completion of the Rewind command.

SECURITY PROTOCOL IN (SPIN) A2h
The SECURITY PROTOCOL IN command (see Table 191) is used to retrieve
security protocol information (see “Security Protocol In Pages (00h - Security
Protocol Information)” on page 202) or the results of one or more SECURITY
PROTOCOL OUT commands (see “SECURITY PROTOCOL OUT (SPOUT) B5h” on
page 215). This command is only supported on Ultrium 4 and later drives.
Table 191. Security Protocol In - A2h CDB
Bit
Byte
7 6 5 4 3 2 1 0
0 Operation code (A2h)
1 Security Protocol
2
Security Protocol Specific
3
4 INC_512
Reserved
(0)
5 Reserved
6 (MSB)
Allocation Length
9 (LSB)
10 Reserved
11 Control Byte (see ″Control Byte Definition″ on page xxx)
The following parameters apply:

v Security Protocol
Table 192. Security Protocol Definitions
Code Description Reference
00h Security protocol information 202
20h Tape Data Encryption 205
all others Reserved
v Security Protocol Specific - The contents depend on the protocol specified by the
Security Protocol field (see Table 192).
v Allocation Length

Security Protocol In Pages (00h - Security Protocol
Information)
The Security Protocol Specific field of the SPIN CDB is defined by Table 193.
Table 193. Security Protocol Specific Definitions for Security Protocol 00h
Code Description Reference
0000h Supported security protocol list See 203
0001h Certificate data 204
0002h - FFFFh Reserved

SPIN (00h:0000h) - Supported Security Protocols List
If the SECURITY PROTOCOL field is set to 00h and the SECURITY PROTOCOL
SPECIFIC field is set to 0000h in a SECURITY PROTOCOL IN command, the
parameter data shall have the format shown in Table 194.
Table 194. Supported Security Protocols List Structure
Bit
Byte
7 6 5 4 3 2 1 0
0
Reserved
5
6 (MSB)
Supported Security Protocol List Length (m-7)
7 (LSB)
8 Supported Security Protocol (first) (00h)
.
.
.
m Supported Security Protocol (last)
The supported Security Protocols are listed in Table 192 on page 201.

SPIN (00h:0001h) - Certificate Data
The drive certificate (if present) is provided in X.509 format via this interface (see
SPC-4).
If the Security Protocol field is set to 00h and the Security Protocol Specific field is
set to 0001h in a Security Protocol In command, the parameter data shall have the
format shown in Table 195.
Table 195. Certificate Data Structure
Bit
Byte 7 6 5 4 3 2 1 0
0
Reserved
1
2 (MSB)
Certificate Length (m-3)
3 (LSB)
4
Certificate
m

v Certificate Length - The total length, in bytes, of the certificate that follows.
v Certificate - The drive certificate set during manufacturing is returned.

Security Protocol In Pages (20h - Tape Data Encryption)
The Security Protocol Specific field (see Table 196) specifies the type of report that
the application client is requesting. Table 196 shows supported Security Protocol
Specific field values.
Page Code Description Reference
0000h Tape Data Encryption In Support Pages see 206
0001h Tape Data Encryption Out Support Pages see 207
0010h Data Encryption Capabilities see 208
0011h Supported Key Formats see 210
0012h Data Encryption Management Capabilities see 211
0020h Data Encryption Status see 212
0021h Next Block Encryption Status see 213

SPIN (20h:0000h) - Tape Data Encryption In Support Pages Page
Supported protocol specific in pages for protocol 20h are indicated above (see
SSC-3).
Table 197. 0000h - Tape Data Encryption In Support Pages Structure
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0000h)
1 (LSB)
2 (MSB)
Page Length (n-3)
3 (LSB)
Tape Data Encryption In Support page code list
4 (MSB)
Tape Data Encryption In Support page code (first)
5 (LSB)
...
n-1 (MSB)
Tape Data Encryption In Support page code (last)
n (LSB)
Table 198 show which Tape Data Encryption In page codes are supported:
Table 198. Tape Data Encryption In page codes
0000h Tape Data Encryption In Support Pages See page 206.
0001h Tape Data Encryption Out Support Pages See page 207.
0010h Data Encryption Capabilities See page 208.
0011h Supported Key Formats See page 210.
0012h Data Encryption Management Capabilities See page 211.
0020h Data Encryption Status See page 212.
0021h Next Block Encryption Status See page213.

SPIN (20h:0001h) - Tape Data Encryption Out Support Pages
Page
Supported protocol specific out pages for protocol 20h are indicated above (see
SSC-3).
Table 199. 0001h - Tape Data Encryption Out Support Pages Structure
Bit
Byte
7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0001h)
1 (LSB)
2 (MSB)
Page Length (n-3)
3 (LSB)
Tape Data Encryption Out Support page code list
4 (MSB)
Tape Data Encryption Out Support page code (first)
5 (LSB)
...
n-1 (MSB)
Tape Data Encryption Out Support page code (first)
n (LSB)
Table 200 shows which Tape Data Encryption Out page codes are supported:
Table 200. Tape Data Encryption Out page codes
0010h Set Data Encryption See page 208.

SPIN (20h:0010h) - Data Encryption Capabilities Page
Table 201 specifies the format of the Data Encryption Capabilities page.
Table 201. 0010h - Data Encryption Capabilities page
Bit
Byte
7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0010h)
1 (LSB)
2 (MSB)
Page Length n-3)
3 (LSB)
4
Reserved
19
Data Encryption Algorithm descriptor list
20
Data Encryption Algorithm descriptor (first)
...
Data Encryption Algorithm descriptor (last)

n

Data Encryption Algorithm Descriptor - Standard Encryption
The Standard Encryption Algorithm Descriptor is shown in Table 202.
Table 202. Data Encryption Algorithm Descriptor - Standard Encryption Structure
Bit
Byte 7 6 5 4 3 2 1 0
0 Algorithm Index (01h)
1 Reserved
2 (MSB)
Descriptor Length (0014h)
3 (LSB)
4 AVFMV SDK_C (0) MAC_C (1) DED_C (1) DECRYPT_C (2H) ENCRYPT_C (2H)
5 Reserved NONCE_C (3) Reserved
6 (MSB) Maximum Unauthenticated Key-Associated Data (U-KAD) Bytes
7 (0020h) (LSB)
8 (MSB) Maximum Authenticated Key-Associated Data (A-KAD) Bytes
9 (000Ch) (LSB)
10 (MSB)
Key Size (0020h)
11 (LSB)
12
Reserved
19
20 (MSB)
Encryption Algorithm Identifier (00000014h)
23 (LSB)

SPIN (20h:0011h) - Supported Key Formats Page
The structure of the Supported Key Formats page is shown in Table 203.
Table 203. Supported Key Formats page Structure
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0011h)
1 (LSB)
2 (MSB)
Page Length (n-3)
3 (LSB)
Supported Key Format list
4 (MSB)
Supported Key Format (first)
5 (LSB)
...
n-1 (MSB)
Supported Key Format (last)
n (LSB)
Table 204. Supported Key Formats

Key Format Description Reference
00h Plaintext Key Format See 210.
Plaintext Key Format (00h): The Plaintext Key Format structure is shown in
Table 205.
Table 205. Plaintext Key Format Structure
Bit
Byte 7 6 5 4 3 2 1 0
o
Key
n

SPIN (20h:0012h) - Data Encryption Management Capabilities
Table 206 specifies the format of the Data Encryption Management Capabilities
page.
Table 206. Data Encryption Management Capabilities
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0012h)
1 (LSB)
2 (MSB)
Page Length (000Ch)
3 (LSB)
4 LOCK_C
Reserved
(1)
5 CKOD_C CKORP_C CKORL_C
Reserved
(1) (1) (1)
6 Reserved
7 AITN_C LOCAL_C PUBLIC_C
Reserved
(1) (1) (1)
8
Reserved
15

SPIN (20h:0020h) - Data Encryption Status Page
Table 207 specifies the format of the Data Encryption Status page.
Note: This is a query of information which was set with Security Protocol Out
0010h - Set Data Encryption and does not reflect the actual state of the
medium itself or of any data on medium.
Table 207. 0020h - Data Encryption Status page
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0020h)
1 (LSB)
2 (MSB)
Page Length (n-3)
3 (LSB)
4 I_T Nexus Scope Reserved Key Scope
5 Encryption Mode
6 Decryption Mode
7 Algorithm Index
8 (MSB)
Key Instance Counter
11 (LSB)
12
Reserved
23
24
Key-Associated Data Descriptors List
n

v I_T Nexus Scope
v Key Scope
v Encryption Mode
v Decryption Mode
v Algorithm Index
v Key Instance Counter
v Key-Associated Data Descriptors List (See “Key-Associated Data (KAD)
Descriptors” on page 219

SPIN (20h:0021h) - Next Block Encryption Status Page
Table 208 specifies the format of the Next Block Encryption Status page.
Table 208. 0021h - Next Block Encryption Status page
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0021h)
1 (LSB)
2 (MSB)
Page Length (n-3)
3 (LSB)
4 (MSB)
Key Instance Counter
11 (LSB)
12 Compression Status (0h) Encryption Status
13 Algorithm Index
14
Reserved
15
16
n

v Logical Object Identifier
v Encryption Status
v Algorithm Index
v Key-Associated Data Descriptors List (See “Key-Associated Data (KAD)
Descriptors” on page 219)
Note: Next block encryption status may not be available in all situations. When it
is not known appropriate values are returned as per the standard. In most
situations next block information is available during read operations when
read ahead is being performed. This is automatically managed by the
device.
The following table indicates valid combinations of record status, Decryption Mode
and returned Key-Associated Descriptors reflecting the currently setup state of the
device.

Table 209. SPIN (20h:0021h) - KAD Parameters by Mode
Key-Associated Descriptors
Record Decryption
Read Data Status aKAD Meta data Notes
Information Mode uKAD 00h Nonce 02h
(DKi) 01h 03h
Unknown any ? 1h n/a n/a n/a n/a
may be
Filemark any n/a 2h n/a n/a n/a n/a2
unknown
may be
EOD any n/a 2h n/a n/a n/a n/a
unknown
may be
Error any n/a 1h n/a n/a n/a n/a
unknown
Cleartext 0h Disable C 3h n/a n/a n/a n/a
not
Cleartext 1h Raw E 3h n/a n/a n/a n/a
readable
not
Cleartext 2h Decrypt E 3h n/a n/a n/a n/a
readable
Cleartext 3h Mixed C 3h n/a n/a n/a n/a
not
Encrypted 0h Disable E 4h Y Y N N
readable
Encrypted 1h Raw R 5h N N N Y
1
Encrypted 2h Decrypt C or Y Y N N
1
Encrypted 3h Mixed C 6h Y Y N N
Legend: Notes:
Y: element is required 1. Data is decrypted
O: element is optional 2. Value may be returned if raw mode and data set is
N: element is not present encrypted
n/a: not applicable (element is not present)
C: cleartext (not encrypted)
R: raw (compressed encoded/encrypted)
E: error condition, record cannot be read
Key-Associated Data (KAD) Descriptors

See “Key-Associated Data (KAD) Descriptors” on page 219.

SECURITY PROTOCOL OUT (SPOUT) B5h
The SECURITY PROTOCOL OUT command (see Table 210) is used to send data to
the logical unit. The data sent specifies one or more operations to be performed by
the logical unit. The format and function of the operations depends on the contents
of the SECURITY PROTOCOL field (see table 178). Depending on the protocol
specified by the SECURITY PROTOCOL field, the application client may use the
SECURITY PROTOCOL IN command (see “SECURITY PROTOCOL IN (SPIN)
A2h” on page 201) to retrieve data derived from these operations. This command
is only supported on Ultrium 4 and later drives.
Note: The operation code, B5h has been recovered from a seldom used media
changer (i.e. LUN 1) command (Request Volume Element Address). If the
device driver being used still uses the LUN field of the CDB from SCSI2
days, this command will be routed to the incorrect LUN. Because it is a
DATA OUT type command, whereas the Request Volume Element Address
is a DATA IN type command this can cause strange system behaviors.
Table 210. Security Protocol Out B5h CDB
Bit
Byte 7 6 5 4 3 2 1 0
0 Operation code (B5h)
1 Security Protocol
2
Security Protocol Specific
3
4 INC_512
Reserved
(0)
5 Reserved
6 (MSB)
Allocation Length
9 (LSB)
10 Reserved
11 Control Bye (see "Control Byte Definition" on page xxx)

v Security Protocol
Value Description
20h Tape Data Encryption security protocol
v Security Protocol Specific - The contents depend on the protocol specified by the
Security protocol field.
v Allocation Length

Security Protocol Out Pages (20h - Tape Data Encryption
Security Protocol)
The Security Protocol Specific field (see Table 210 on page 215) specifies the type of
page that the application client is sending. Table 211 shows supported values.
0010h Set Data Encryption See 208
SPOUT (20h:0010h) - Set Data Encryption

Table 212 specifies the format of the Set Data Encryption page.
Table 212. 0010h - Set Data Encryption page
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
Page Code (0010h)
1 (LSB)
2 (MSB)
Page Length (m-3)
3 (LSB)
4 Scope Reserved LOCK
5 Reserved SDK (0) CKOD CKORP CKORL
6 Encryption Mode
7 Decryption Mode
8 Algorithm Index
9 Key Format
10
Reserved
17
18 (MSB)
Key Length (n-19)
19 (LSB)
20
Key
n
n+1
m

v Scope
Value Description
0h Public
1h Local
2h All I_T Nexus
v LOCK
v CKOD
v CKORP
v CKORL

v Encryption Mode
Value Description
0h Disable
1h External
2h Encrypt
v Decryption Mode
Value Description
0h Disable
1h Raw
2h Decrypt
3h Mixed
v Key Format
v Key Length
Value Description
000h When no Key is specified
0020h When Key is specified using Key Format 00h
v Key
v Key-Associated Descriptors List (See “Key-Associated Data (KAD) Descriptors”
on page 219)
The following table indicates valid combinations of Encryption Mode and

Decryption Mode and mandatory, optional and prohibited Key and Key-Associated
Descriptors.

Table 213. SPOUT (20h:0010h) - KAD Parameters by Mode
Key-Associated Descriptors
Encryption Decryption aKAD
R/W Data Key uKAD Nonce Meta data Notes
Mode Mode (DKi)
00h 02h 03h
01h
0h Disable 0h Disable C/C P P P P P
0h Disable 1h Raw R/C P P P P P not recommended
4 2
0h Disable 2h Decrypt C /C M P P P P not recommended
6 2
0h Disable 3h Mixed C /C M P P P P
1h External 0h Disable C/R P P P P M1 not recommended
1
1h External 1h Raw R/R P P P P M
4 2
1h External 2h Decrypt C /R M P P P M1 not recommended
6 2 1
1h External 3h Mixed C /R M P P P M not recommended
4 1 1 1 1,3
2h Encrypt 0h Disable C/C M O O O P
4 1 1 1 1,3
2h Encrypt 1h Raw R/C M O O O P not recommended
5 4 1 1 1,3
2h Encrypt 2h Decrypt C /C M O O O P
6 4 1 1 1,3
2h Encrypt 3h Mixed C /C M O O O P
Legend: Notes:
M: element is mandatory (required) 1: Only used for writing
P: element prohibited (must not be present) 2: Only used for reading
O: element is optional (may be device 3: May be partially ignored
generated) 4: Data is encrypted
I: element is ignored (may be present) 5: Data is decrypted
C: cleartext (not encrypted) 6: Data is decrypted (if needed)
R: raw (compressed encoded/encrypted)

Key-Associated Data (KAD) Descriptors
KAD 00h - Unauthenticated KAD
The Unauthenticated KAD field is available for application use and is recorded
when writing with each key change.
Note: If this field is not present an identifier of all zeroes is used.

Table 214. KAD 00h - Unauthenticated KAD
Bit
7
Byte msb 6 5 4 3 2 1 0
0 Key Descriptor Type (00h)
1 Reserved Authenticated
2 (MSB)
Key Descriptor Length (n-3)
3 (LSB)
4
Unauthenticated Data
n

v Authenticated
Value Description
0h Reserved (must be set for Security Protocol Out - 0010h)
1h Not Covered by Authentication (only Security Protocol In)
v Key Descriptor Length: may be up to 0020h bytes
v Unauthenticated Data

KAD 01h - DKi (Data Key Identifier)
The DKi KAD field is an optional field which is used when writing and is
recorded with each record.
Note: If this field is not present an identifier of all zeroes is used.

Table 215. KAD 01h - DKi (Data Key Identifier)
Bit
7 0
Byte msb 6 5 4 3 2 1 lsb
2 (MSB)
3 (LSB)
4
DKi
n

v Authenticated
Value Description
0h Reserved (must be set for Security Protocol Out)
2h No attempt has been made to authenticate (only Security Protocol In)
v Key Descriptor Length: may be up to 000Ch bytes
v DKi

KAD 02h - Nonce
The Nonce/Initialization Vector (IV) is not technically key-associated data. The
Nonce may be set to provide the initial value for IV generation for write
operations. This field is optional and the device is capable of generating high
quality random IV values. When an application specifies nonce values, it is
possible that a systemic cryptographic weakness may be introduced into the
system. It is strongly recommended that nonce values are not supplied by the
application.
Note: The Nonce KAD is only reported by the device in Security Protocol In 0020h
- Data Encryption Status, and the value returned is the exact value specified
in 0010h - Set Data Encryption. This may not reflect the actual nonce or IV
used for writing encrypted data.
Note: IV values start at the specified Nonce value and are incremented for each
record written.
Table 216. KAD 02h - Nonce
Bit
7 0
2 (MSB)
Key Descriptor Length (n-3) (000Ch)
3 (LSB)
4
Nonce/IV
n

v Authenticated
Value Description
v Key Descriptor Length: the only supported length is 000Ch
v Nonce/IV

KAD 03h - Metadata (Raw)
The Metadata KAD is the combination of the full DKi and DKv fields. This
information is available during raw decryption mode reads and must be
propagated during external encryption mode write operations (keyless copy) of
tape.
Table 217. KAD 03h - Metadata (Raw)
Bit
7 0
2 (MSB)
3 (LSB)
4
Metadata
n

v Authenticated
Value Description
v Key Descriptor Length: the only supported length is 0040h
v Metadata

SEND DIAGNOSTIC
Table 218. SEND DIAGNOSTIC Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Self-Test Code (000b) PF Reserved SelfTest DevOfL UnitOfL
2 Reserved
3
4
5 Control
The SEND DIAGNOSTIC command requests the drive to perform diagnostic

operations. When the SelfTest field is 0 and the Self-Test Code field contains 000b,
this command is usually followed by a RECEIVE DIAGNOSTIC RESULTS
command.
Table 219 lists the supported diagnostics.

Table 219. Supported Diagnostics
Name Diagnostic ID See Page...
Self-Test N/A See page230.
Post A Self Test Diagnostic 0100h See page 231.
Post B Performance Diagnostic 0101h See page232.
Post C Media Test Diagnostic 0102h See page234.
Post D Head Test Diagnostic 0103h See page235.
Primary Port Wrap Test 0090h See page236.
Force Dump 0160h See page237.
Write Dump to Cartridge 0161h See page238.
Write Dump to FLASH (EEPROM) 0162h See page241.
Clear FLASH Dump (EEPROM) 016Fh See page241.
Set Traps 0190h See page242.
Remove Traps 0191h See page243.
Reset Drive 2002h See page245.
Read Thermal Sensor 1002h See page245.

SIM Data Structure
The following data structure is used in several of the diagnostics. Its purpose is to
give detailed error information about drive problems:
Table 220. SIM Data Structure
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (31h)
1 Reserved
2
: Page Length (0044h)
3
4
: Parameter Code (0000h)
5
6 Parameter Control (061h)
7 Parameter Length (040h)
8 Indicator (01h)
9
: Reserved
15
16
: Microcode Level
19
20
: SIM Message Code
21
22
: Reserved
23
24 Exception Message
25 Service Message
26 Severity Code
27 Reserved
28
: Exception Data (00h)
29

Table 220. SIM Data Structure (continued)
Bit
Byte 7 6 5 4 3 2 1 0
30
: Error Code
33
34
: First FSC
37
38
: Last FSC
41
42
: Product ID (8000h)
45
46
: Vendor ID “IBM”
48
49
: Plant of Manufacture
50
51 Product ID3 ’-’
52
: Serial Number
63
64
: Device Type/Dev SIM_MESSAGE_TYPE
71
The SIM Message Code may be one of six values:

v ’00’=No Message
v ’41’=Device Degraded
v ’42’=Device Hardware Failure
v ’43’=Service Circuit Failed
v ’55’=Drive Needs Cleaning
v ’57’=Drive Has Been Cleaned
The Exception Message may be one of eight values:

v ’1’=Effect of Failure is Unknown

v ’2’=Device Exception No Performance Impact
v ’3’=Exception on SCSI Interface xx
v ’4’=Device Exception on Operator Panel
v ’5’=Device Exception on Tape Path
v ’6’=Device Exception in Drive
v ’7’=Cleaning Required
v ’8’=Cleaning Done
The Service Message may be one of four values:

v ’1’=Repair Impact is Unknown
v ’7’=Repair will Disable Access to Device Servo
v ’9’=Clean Device
v ’A’=Device Cleaned
The Severity Code may be one of four values:

v ’0’=Service
v ’1’=Moderate
v ’2’=Serious
v ’3’=Acute

MIM Data Structure
The following data structure is used in several of the diagnostics. Its purpose is to
give detailed error information about media problems:
Table 221. MIM Data Structure
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (31h)
1 Reserved
2
3
4
: Parameter Code (0000h)
5
6 Parameter Control (061h)
7 Parameter Length (040h)
8 Indicator (02h)
9
: Reserved
15
16
: Engineering Data Microcode Level
19
20
: MIM Message Code
21
22
: SARS Data
23
24 Exception Message
25 Reserved
26 Severity Code
27
: Reserved
29
30
: Error Code
33

Table 221. MIM Data Structure (continued)
Bit
Byte 7 6 5 4 3 2 1 0
34
: Volume ID Volume Serial Number
39
40 Volume ID Flag
41 Reserved
42
: Product ID (8000h)
45
46
: Vendor ID “IBM”
48
49
: Plant of Manufacture
50
51 Product ID3 ’-’
52
: Serial Number
63
64
: Device Type
71
The MIM Message Code may be one of four values:

v ’00’=No Message
v ’60’=Bad Media Read Only Permitted
v ’61’=Rewrite Media if Possible
v ’72’=Replace Cleaning Cartridge
The Exception Message may be one of four values:

v ’2’=Data Degraded
v ’4’=Medium Degraded
v ’6’=CM Error
v ’7’=Medium Exception
The Severity Code may be one of four values:

v ’0’=Service
v ’1’=Moderate - Temporary Read/Write Errors
v ’2’=Serious - Permanent Read/Write Errors
v ’3’=Acute - CM Error
The Volume ID Flag may be one of four values:

v ’0’=VOLID not valid
v ’1’=VOLID valid - obtained from tape (CM)
v ’3’=VOLID valid - obtained from cartridge label (server data)
v ’5’=VOLID valid - obtained from cartridge level (library)

Self-Test
For the Self-Test Diagnostic, the CDB values must be set as follows:
v PF - Any value allowed and ignored
v SelfTest - 1
v DevOfl - Any value allowed and ignored
v UnitOfl - Any value allowed and ignored
v Parameter List Length - 0000h
Receive Diagnostics Results: There are no diagnostic results for the Self-Test
diagnostic.

Post A Self Test Diagnostic
Table 222. Post A Self Test Send Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
: Diagnostic ID (0100h)
5
Cartridge
6 Flags (0000000b) Required
(bx)
7 Reserved
Table 223. Post A Self Test Receive Diagnostic Parameter Data

Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
: Page Length (0004Dh)
3
4
5
6 Flags
7 Reserved
Diag. SIM/MIM Error
8 Flags (00000b)
Blocked Present
9
: SIM/MIM Message or All Zeros
80

Post B Performance Diagnostic
The Performance Diagnostic performs a test to determine how well the tape drive
writes data. If the percentage degradation exceeds the threshold, the Send
Diagnostic command will return a Check Condition. The Sense Key will be set to
1h and the ASC/ASCQ set to 0000h. The FSC field of sense data will be set to
52E5, if the degradation is in the forward direction, or 52E6, if the degradation was
in the backward direction. The Flag field of the sense data will be set to the
percentage. TapeAlert2 (Write Warning) will be set and the SCD will display the
character ’A.’ The Receive Diagnostic command will return SIM data related to the
failure.
Table 224. Post B Performance Send Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(1b)
7 Reserved
Table 225. Post B Performance Receive Diagnostic Parameter Data

Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
6 Flags
7 Reserved
Diag. SIM/MIM Error
8 Flags (00000b)
Blocked Present

Table 225. Post B Performance Receive Diagnostic Parameter Data (continued)
Bit
Byte 7 6 5 4 3 2 1 0
9
80

Post C Media Test Diagnostic
Table 226. Post C Media Test Send Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(1b)
7 Reserved
Table 227. Post C Media Test Receive Diagnostic Parameter Data

Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
6 Flags
7 Reserved
Diag. SIM/MIM Error
8 Flags (00000b)
Blocked Present
9
80

Post D Head Test Diagnostic
Table 228. Post D Head Test Send Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(1b)
7 Reserved
Table 229. Post D Head Test Receive Diagnostic Parameter Data

Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
6 Flags
7 Reserved
Diag. SIM/MIM
8 Flags (00000b) Error
Blocked Present
9
80

Primary Port Wrap Test Diagnostic
This test will perform a wrap test on the specified primary port. A wrap tool must
be attached prior to running this command.
See Table 230.

Table 230. Primary Port Wrap Test Send Diagnostic Parameter Data
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2 (MSB)
Page Length (0004h)
3 (LSB)
4
Diagnostic ID (0090h)
5
Cartridge
(0b)
7 Port Identifier
Port Identifier
This field is synonymous with the Port Identifier field described in the Device
Identification Inquiry page. If the value of the Port Identifier is zero, then the wrap
test will be performed on all primary ports. If the value is a valid Port Identifier,
the wrap test is performed on the port indicated by the Port Identifier. If an invalid
Port Identifier value is used, the drive will respond with a Check Condition for
Invalid Field in Parameter List.
The Primary Port Wrap Test Receive Diagnostic returns information about the
attempted Primary Port Wrap Test Send Diagnostic.
Table 231. Primary Port Wrap Test Receive Diagnostic Parameter Data
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2 (MSB)
Page Length (004Dh)
3 (LSB)
4
5
6 Flags
7 Reserved
Diag. SIM/MIM
Blocked Present
9
SIM/MIM Message or All Zeros
80

Force Dump
The Force Dump Diagnostic causes the drive to capture into a reserved space in
memory a log of data that is used to assist field and development engineers in
determining the root cause of drive behavior. This log is called a drive dump. This
diagnostic copies the dump to memory, but does not save it for later use. The save
operation must be done by using a READ BUFFER command or a Write Dump To
Cartridge Diagnostic.
Note: Because forcing a drive dump will overwrite any previously stored dump,
before forcing the dump it may be desirable to check the Dump field of the
sense data to determine if a drive dump exists.
For the Force Dump Diagnostic, the CDB values should be set as follows:
v PF - 1
v SelfTest - 0
Table 232 shows the parameter data to be sent with the Force Dump diagnostic.
Table 232. Force Dump Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(0)
7 Reserved
Receive Diagnostics Results: There are no diagnostic results for the Force Dump
diagnostic.

Write Dump To Cartridge
The Write Dump To Cartridge Send Diagnostic causes the drive to write a dump
from memory to the cartridge that is loaded in the drive. The dump might exist
because of a previous Force Dump diagnostic command or it might exist because
of other scenarios where the drive code automatically creates a dump. To
determine if a drive dump exists before forcing a dump, the application client can
look at the Dump field of the sense data.
Table 233. Write Dump to Cartridge Send Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(1b)
7 Reserved
The Write Dump To Cartridge Receive Diagnostic returns information about the
attempted Write Dump To Cartridge Send Diagnostic.

Table 234. Write Dump to Cartridge Receive Diagnostic Parameter Data
Bit
Byte
7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
6 Flags
7 Reserved
Diag. SIM/MIM
Blocked Present
9
80

Write Dump to Flash (EEPROM)
The Write Dump To FLASH Send Diagnostic causes the drive to write a dump
from memory to non-volatile memory in the drive. If a dump exists before receipt
of this command that dump is written to non-volatile memory. If a dump does not
exist before receipt of this command, a dump will be forced then written to
non-volatile memory. The intent of this function is to help the factory in
diagnostics of a failed drive in the field. This dump will be returned to a series of
read buffer commands as any normal dump will be read. This dump will only be
returned if no other dump exists in the drive.
See Table 235.

Table 235. Write Dump To FLASH Send Diagnostic Parameter Data
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2 (MSB)
Page Length (0004h)
3 (LSB)
4
5
Cartridge
(0b)
7 Reserved
The Write Dump To FLASH Receive Diagnostic returns information about the
attempted Write Dump To FLASH Send Diagnostic.
See Table 236.

Table 236. Write Dump To FLASH Receive
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2 (MSB)
Page Length (004Dh)
3 (LSB)
4
Diagnostic ID (162hh)
5
6 Flags
7 Reserved
Diag. SIM/MIM
Blocked Present
9
SIM/MIM Message or All Zeros
80

Clear FLASH Dump (EEPROM)
The Clear FLASH Dump Send Diagnostic causes the drive to clear the dump that
was saved to non-volatile memory in the drive. The intent of this function is to
clear a dump in non-volatile memory that is no longer needed for factory
diagnostics.
See Table 237

Table 237. Clear Dump to FLASH Send Diagnostic Parameter Data
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2 (MSB)
Page Length (0004h)
3 (LSB)
4
Diagnostic ID (016Fh)
5
Cartridge
(0b)
7 Reserved

Set Traps
The Set Traps diagnostic is used to cause the drive to force a Panic when the
specified Fault Symptom Code (FSC) is created in the drive. A Panic will cause the
drive to capture a drive dump, then reboot. The drive dump will be available for
retrieval after the reboot.
In Ultrium 2 and later drives, a list of traps can be created. When the list is full,
the next trap that is set will automatically clear the least recently set trap.
Currently, the maximum number of traps that can be set is 10, but may be
modified at any time.
In Ultrium 1 drives only one trap is allowed to be set at a time.
For the Set Traps Diagnostic, the CDB values should be set as follows:
v PF - 1
v SelfTest - 0
v Parameter List Length - 000Ah
Table 238 shows the parameter data to be sent with the Set Traps diagnostic.
Table 238. Set Traps Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(0)
7 Reserved
8
: Fault Symptom Code
Receive Diagnostics Results: There are no diagnostic results for the Set Traps
diagnostic.

Remove Traps
The Remove Traps diagnostic is used to remove a trap that has been previously set
by a Set Traps diagnostic. The Fault Symptom Code (FSC) that is specified is
cleared from the trap list.
For the Remove Traps Diagnostic, the CDB values should be set as follows:
v PF - 1
v SelfTest - 0
v Parameter List Length - 000Ah
Table 239 shows the parameter data to be sent with the Remove Traps diagnostic.
If all that is desired is to inspect which Traps have been set, sending down Remove
Traps with the Fault Symptom Code set to 0000h will set up diagnostic results that
show which traps are currently set without changing the state of any traps.
Table 239. Remove Traps Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(0)
7 Reserved
8
: Fault Symptom Code
Receive Diagnostics Results - Table 240 on page 244 shows the diagnostic results
data returned for the Remove Traps diagnostic.

Table 240. Reset Drive Diagnostic Results Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
: Page Length (n - 4)
3
4
5
6 Flags (0)
7 Reserved
Diag Reserved Error
8 Reserved
Blocked
9-10
First Fault Symptom Code with Trap Still Set
:
Last Fault Symptom Code with Trap Still Set
(n-1)-n

Reset Drive
The Reset Drive diagnostic causes the drive to reboot. All data in the drive is lost.
For the Reset Drive Diagnostic, the CDB values should be set as follows:
v PF - 1
v SelfTest - 0
v DevOfl - 1
Table 241 shows the parameter data to be sent with the Reset Drive.
Table 241. Reset Drive Diagnostic Parameter Data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2
3
4
5
Cartridge
(0)
7 ReservedXRE
Receive Diagnostics Results: There are no diagnostic results for the Reset Drive
diagnostic.
Read Thermal Sensor

The Read Thermal Sensor diagnostic is used to read the digital thermal sensor on
the drive. For the Read Thermal Sensor, the CDB values should be set as follows:
v PF - 1
v SelfTest - 0
Table 242 on page 246 shows the parameter data to be sent with the Read Thermal
Sensor diagnostic.

Table 242. Read Thermal Sensor Diagnostic parameter data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2 (MSB)
Page Length (0004h)
3 (LSB)
4 (MSB)
5 (LSB)
Cartridge
(0b)
7 Reserved
Receive Diagnostics Results - Table 243 on page 247 shows the diagnostic results
data returned for the Read Thermal Sensor diagnostic.

Table 243. Read Thermal Sensor Receive Diagnostic parameter data
Bit
Byte 7 6 5 4 3 2 1 0
0 Page Code (80h)
1 Reserved
2 (MSB)
Page Length (000Eh)
3 (LSB)
4 (MSB)
5 (LSB)
Cartridge
(0b)
7 Reserved
Diag.
8 Flags (00000b) Reserved Error
Blocked
9 Reserved
Thermal Reading (units of oC)
10
(Value read by sensor)
11 Reserved
Thermal Warning Threshold (units of oC)
12
(Full-Height Drives raise an alert at this temperature)
13 Reserved
Thermal Fencing Threshold (units of oC)
14
(Half-Height Drives fence operations at this temperature)
15 Reserved
Thermal Restart Threshold (units of oC)
16
(Half-Height Drives remove the temperature fence at this temperature)
17 Reserved
Note: A value of 00h in any of the Thermal value fields indicates the value is not applicable.

SET CAPACITY (not Ultrium 1)
The SET CAPACITY command is supported in Ultrium 2 and later drives only and
and sets the available medium for the currently mounted tape to a proportion of
the total capacity of that tape. Any excess space will be unavailable on the tape
after successful completion of this command until changed by a new SET
CAPACITY command. This change will persist through power cycles, logical unit
resets, and the unloading or reloading of the tape.
Table 244. SET CAPACITY Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Reserved Immed
2 Reserved
3 (MSB)
: Capacity Proportion Value
4 (LSB)
5 Control
If the tape drive does not contain a tape, then the command will be terminated
with Check Condition status. The Sense Key must be set to Not Ready, and the
Additional Sense Code must be set to Medium Not Present.
The SET CAPACITY command will be accepted only when the tape is at
beginning-of-tape (BOT). If the tape is logically at any other position, the command
will be rejected with Check Condition status. The Sense Key will be Illegal
Request, and the Additional Sense Code set to Position Past Beginning Of Medium
(5/3B0Ch).
A valid SET CAPACITY command will cause all data on the entire physical tape to
be lost.
Buffered write data may be discarded by the tape drive upon successful validation
of the SET CAPACITY command.
An Immediate (Immed) field of 0 specifies that the device server will not return
status until the set capacity operation has completed. An Immed field of 1 specifies
that the device server will return status as soon as the command descriptor block
of the SET CAPACITY command has been validated. If Check Condition status is
returned for a SET CAPACITY command with an Immed field set to 1, the set
capacity operation has not been performed.
The Capacity Proportion Value field specifies the portion of the total unscaled tape
capacity to be made available for use. This field is the numerator to a fraction that
has a denominator of 65 535 (FFFFh). The resulting available capacity on the tape
is equal to the total unscaled tape capacity multiplied by this fraction.
((Capacity Proportion Value) / 65 535) * (Total Unscaled Tape Capacity) =

(Resulting Available Capacity)

The tape drive may round up the capacity to the next highest supported value.
This rounding is not considered an error and will not be reported. On Ultrium 2
RoHS, Ultrium 3 and later drives, if the Capacity Proportion Value does not meet
the ranges specified in Table 194 then Check Condition status is returned with
Illegal Field in CDB (5/2400h).
On Ultrium 2 non-RoHS drives, if the Capacity Proportion Value does not meet the
ranges specified in Table 194 or it attempts to increase the available capacity to a
value higher than it is currently set then Check Condition status is returned with
Illegal Field in CDB (5/2400h).
WARNING: If the Capacity Proportion is increased it is possible that a debris

dump may exist at the previous logical end of medium. This may cause a defect
when attempting to write through that area.
Table 245. Minimum Capacity Proportion Values
Minimum Capacity Resultant Approximate
Cartridge Maximum Capacity
Proportion Value Minimum Capacity
Ultrium 2 1605h 17.2 GB 200 GB
Ultrium 3 151Ah 33 GB 400 GB
Ultrium 4 123Dh 52 GB 800 GB
Note: Available and total tape capacities are approximate values that may be affected by defects which reduce the
actual available capacity of the tape. Other factors, such as compression and block packing, may also affect available
capacity.

SET TIMESTAMP (Not LTO 1)
The SET TIMESTAMP command (see Table 246) requests the drive to initialize the
timestamp (see “Device Clocks” on page 265), if the SCSIP bit is set to one or the
TCMOS bit is set to one in the Control Extension mode page. If the SCSIP bit is set
to zero, the SET TIMESTAMP command is terminated with CHECK CONDITION
status, with the sense key set to ILLEGAL REQUEST, and the additional sense
code set to INVALID FIELD IN CDB.
Table 246. SET TIMESTAMP command
Bit
Byte 7 6 5 4 3 2 1 0
0 OPERATION CODE (A4h)
1 Reserved SERVICE ACTION (TBD)
2
Reserved
5
6
PARAMETER LIST LENGTH
9
10 Reserved
11 CONTROL
The PARAMETER LIST LENGTH field specifies the length in bytes of the SET
TIMESTAMP parameters that is transferred from the application client to the
device server. A parameter list length of zero indicates that no data is transferred,
and that no change is made to the timestamp.
The format for the parameter data returned by the SET TIMESTAMP command is
shown in Table 247.
Table 247. SET TIMESTAMP parameter list format
Bit
Byte 7 6 5 4 3 2 1 0
0
Reserved
1
2 Reserved
3 Reserved
4 (MSB)
TIMESTAMP
9 (LSB)
10 Reserved
11 Reserved
The TIMESTAMP field contains the initial value of the timestamp in the format
defined in “Device Clocks” on page 265. The timestamp should be the number of
milliseconds that have elapsed since midnight, 1 January 1970 UT. If the high order
byte in the TIMESTAMP field is greater than F0h, the command is terminated with
CHECK CONDITION status, with the sense key set to ILLEGAL REQUEST, and
the additional sense code set to INVALID FIELD IN PARAMETER LIST.

On successful completion of a SET TIMESTAMP command the drive generates a
unit attention condition for the initiator port associated with every I_T nexus
except the I_T nexus on which the SET TIMESTAMP command was received, with
the additional sense code set to TIMESTAMP CHANGED.

SPACE
The SPACE (6) command instructs the drive to set a new logical position relative
to the current logical position. How this is done depends on the value of the Code
field and the Count field. The Count field is a signed value that indicates the
distance to move. A negative value indicates movement towards BOM; a positive
value indicates movement towards EOM.
Table 248. SPACE (6) Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved Code
2
: Count
4
5 Control
Table 249. SPACE (16) command

Bit Byte 7 6 5 4 3 2 1 0
0 Operation Code (91 h)
1 Reserved Code
2 Reserved
3 Reserved
4 (MSB)
5
6
7
8 Count
9
10
11 (LSB)
12 (MSB)
Parameter Length (0000h)
13 (LSB)
14 Reserved
15 Control
If the Code field is 000b, then the logical position is moved the number of blocks
that is indicated by the Count field. If a filemark is encountered while spacing over
blocks, the command is terminated. Check Condition status is returned, and the
Filemark and Valid fields are set to 1 in the sense data. The Sense Key is set to No
Sense and the Additional Sense Code is set to Filemark Detected. The Information
field is set to the requested count minus the actual number of blocks spaced over
(not including the filemark). The new logical position is set immediately after the
file mark in the direction of the space operation. If BOM or EOD is detected before
the requested logical position, then the logical position is set to that position.

If the Code field is 001b, then the logical position is moved the number of file
marks indicated by the Count field. If BOM or EOD is detected before the
requested logical position, then the logical position is set to that position.
If end-of-data is encountered while spacing over blocks or filemarks, Check

Condition status is returned, the Sense Key is set to Blank Check, and the sense
data Valid field is set to 1 in the sense data. The Additional Sense Code is set to
End-Of-Data Detected. The sense data EOM field is set to 1 if end-of-data is
encountered at or after early-warning. The Information field is set to the requested
count minus the actual number of blocks or filemarks spaced over as defined by
the Code value. The medium is positioned such that a subsequent write operation
would append to the last record or filemark.
If the end-of-tape is encountered while spacing forward over blocks or filemarks,

Check Condition status is returned, and the Sense Key is set to Medium Error. The
Additional Sense Code is set to End-Of-Partition/Medium Detected, and the sense
data EOM and Valid fields are set to 1. The Information field is set to the requested
count minus the actual number of blocks or filemarks spaced over, as defined by
the Code value.
If beginning-of-tape is encountered while spacing over blocks or filemarks in the

reverse direction, the drive returns Check Condition status and sets the Sense Key
to No Sense. The Additional Sense Code is set to Beginning-Of-Partition/Medium
Detected. The sense data EOM and Valid fields are set to 1, and the Information
field is set to the total number of blocks or filemarks not spaced over (the
requested number of blocks or filemarks minus the actual number of blocks or
filemarks spaced over). A successfully completed SPACE command does not set
EOM to 1 at beginning-of-tape.
If the Code field is 011b, then the logical position is set to after the last valid block
on tape. In this case the Count field is ignored.
Any other value of the Code field causes Check Condition status to be returned.
Spacing to sequential file marks is not supported. Set marks are not supported. The
Sense Key is set to Illegal Request and the ASC/ASCQ is set to Invalid Field in
CDB (2400).
The explicit command set is not supported. The Parameter Length field of the
SPACE (16) command is set to 0. If the Parameter Length field is set to any other
value, the command is terminated with a Check Condition status. The sense key is
set to Illegal Request and the additional sense code is set to Invalid Field In Cdb
(5/2400h)

TEST UNIT READY
The TEST UNIT READY command reports when there is a medium in the drive
that is ready to accept medium access commands. If a medium is loaded and a
problem is detected that does not force the automatic ejection of the medium,
GOOD status will still be returned to the TEST UNIT READY command. Whatever
error is detected will be returned in response to the next medium access command.
Table 250. TEST UNIT READY Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved
2 Reserved
3 Reserved
4 Reserved
5 Control

VERIFY
The VERIFY command causes data to be read from the tape and passed through
the drive’s error detection and correction hardware to determine whether it can be
recovered from the tape. The amount of data to be read is indicated by the
Verification Length field and the Fixed field in the same manner as is used in a
READ command. (See “READ” on page 154.)
Table 251. VERIFY Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved Immed(0) BCmp(0) Fixed
2
: Verification Length
4
5 Control
The VERIFY command is supported by all drives.
The Immed and BCmp fields are not supported and must be set to 0.

WRITE
The WRITE command causes data to be transferred to the drive in a Data Out and
written to tape.
Table 252. WRITE Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved Fixed
2
: Transfer Length
4
5 Control
If the Fixed field is set to 1, the Block Length (see “Mode Block Descriptor” on
page 112) is set to 0, and the Transfer Length field is not 0, Check Condition status
is returned with Illegal Field in CDB (5/2400h).
If the Fixed field is set to 0, the initiator transfers a single block of the length
indicated in Transfer Length.
If the Fixed field is set to 1, the initiator transfers a sequence of blocks. The
number of blocks is given by the Transfer Length field. The length of the blocks is
given by the current fixed block length (see “Mode Block Descriptor” on page 112).
If the current logical block number is greater than FFFFFF00h and less than
FFFFFFF0h, rules for Logical EOM processing are applied. If the current logical
block number is greater than or equal to FFFFFFF0h, rules for physical end of tape
processing are applied.

WRITE ATTRIBUTE
Table 253. WRITE ATTRIBUTE Command
Bit
Byte 7 6 5 4 3 2 1 0
1
: Reserved
4
5 Volume Number (0)
6 Reserved
8
: Reserved
9
10
: Allocation Length
13
14 Reserved
15 Control
Refer to SCSI Primary Commands-3 (SPC-3) for support for the WRITE ATTRIBUTE
command.
For information about attributes that are supported, see “READ ATTRIBUTE” on
page 156.
Table 254 gives the format of the data that is returned for an Attribute Values
service action request.
Table 254. Parameter Data for Attribute Values Service Action Request
Bit
Byte 7 6 5 4 3 2 1 0
0
: Parameter Data Length (n-3)
3
4
: Attribute #1
x
m
: Attribute #y

WRITE BUFFER
The WRITE BUFFER command transfers data into the memory on the drive for the
purpose of diagnostics, tests, or firmware upgrade.
Table 255. WRITE BUFFER Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Mode
2 Buffer ID
3
: Buffer Offset
5
6
8
9 Control
Values for the Mode field and their meaning are described in Table 256.
Table 256. Description of Mode Field
Mode Description Support
00h Write combined header and data 1, 2, 3, 4
02h Write data 1, 2, 3, 4
04h Download microcode 1, 2, 3, 4
05h Download microcode and save 1, 2, 3, 4
06h Download microcode with offsets 1, 2, 3, 4
07h Download microcode with offsets and save 1, 2, 3, 4
0Ah Echo buffer 2, 3, 4
Legend:
If the Mode field is 00h, the data to be written is sent with a 4-byte header, which
must be set to all zeroes.
The Buffer ID field indicates which buffer is to be written.
To download firmware, Modes 04h, 05h, 06h, and 07h are accepted and handled in
the same fashion. Any Buffer ID value in these modes is allowed and ignored. The
code must be downloaded with strictly increasing offsets. If it is not, no data is

written and Check Condition status is generated. The Sense Key is set to Illegal
Request (5) and the ASC/ASCQ is set to Invalid Field in CDB (2400). Ending status
is given when the code has been successfully saved to flash, but before the drive
has initiated its reset.
If Mode is set to 0Ah, the data is stored in the echo buffer. The Buffer ID and
Buffer Offset fields are ignored in this mode.
The Buffer Offset field indicates where in the buffer the data should be written.
This must be smaller than the size of the buffer.
The Parameter List Length field holds the amount of data. This must be smaller
than the difference between the Buffer Offset field and the size of the buffer. If it is
not, no data is written and Check Condition status is generated. The Sense Key is
set to Illegal Request (5) and the ASC/ASCQ is set to Invalid Field in CDB (2400).
Buffers that may be written to are Test, Firmware, and VPD (see “READ BUFFER”
on page 173).

WRITE FILE MARKS
The WRITE FILE MARKS command causes a sequence of file marks to be written
at the current logical position. The number of file marks to be written is indicated
in the Count field. If the Immed field is set, status is returned immediately, before
the file marks are written to tape. If the Immed field is set to 0, the file marks and
any buffered data is written to tape before status is returned.
The Write Set Mark (WSmk) field must be 0. Set marks are not supported.
Table 257. WRITE FILE MARKS Command
Bit
Byte 7 6 5 4 3 2 1 0
1 Obsolete Reserved WSmk(0) Immed
2
: Count
4
5 Control
If the Immed field is set to 0 and the Count field is 0, then all buffered data is
flushed to tape before the status is reported.
If the current logical block number is greater than FFFFFF00h and less than
FFFFFFF0h, rules for Logical EOM processing are applied. If the current logical
block number is greater than or equal to FFFFFFF0h, rules for physical end of tape
processing are applied.
A Write Filemark with Count set to 0 when there is no data in the buffer to be
synchronized to media will always return Good Status. Good Status is returned
even if the media is WORM media and the media is not at a writable location or
the WORM media has been tampered with.
A Write Filemark with Count set to 0 when there is data in the buffer to be
synchronized with media will return sense data as expected for a Write.
On eServer enabled drives, Bit 7 (the MSB of the vendor specific field) of the
control byte is ignored.

Chapter 4. Error Sense Information
Sense Data
For a description of Sense data, see “Sense Data Format” on page 194.
Sense Data Management

The drive maintains three types of Sense data:
Current Sense
The Sense data associated with the last command received from the
initiator.
Deferred Sense
The Sense data from a command that has been reported as Good, but has
generated sense data after being reported. This may be a command with
the Immediate flag set or may be a buffered write. A command with the
Immediate flag set generates sense for the server that sent the command. A
buffered write may generate sense for all servers.
Unit Attention Sense
The Sense data generated by a Unit Attention condition. (See “Unit
Attention Conditions” on page 261.) This is generated for all servers. When
a Unit Attention condition has been driven by a command (for example,
when mode parameters have changed), a Unit Attention is posted for all
initiators except the one that caused the change.
Commands generate Check Condition status if there is Deferred Sense data or Unit
Attention data available, depending on Table 8 on page 17. All commands generate
Check Condition status if the command itself generates sense data. If the next
command after the Check Condition status is not a REQUEST SENSE command,
then all sense data for that initiator is cleared.
When a REQUEST SENSE command is received, the Current Sense is returned. If

there is no Current Sense, the Deferred Sense is returned. If there is no Deferred
Sense, the Unit Attention Sense is returned. If there is no Unit Attention Sense,
default sense data is returned. Once a particular set of sense data has been
returned, that sense data is cleared. Any other sense data that is still pending may
still cause Check Condition status for subsequent commands.
Unit Attention Conditions

The drive generates a Unit Attention condition under the following circumstances:
v Reset condition (for example, power-on, SCSI reset, bus device reset)
v Tape Loaded condition (for example, media inserted, LOAD command from
another initiator)
v Mode parameters changed by another initiator
v Drive firmware has been upgraded
The drive only maintains one Unit Attention condition at any one time for any one
initiator. If a subsequent Unit Attention condition is generated, it replaces the

existing one if it is of higher priority. If it is of lower priority, it is ignored. The
priorities are in the order listed above, with a reset being highest priority and a
firmware upgrade being lowest priority.
Persistent Errors
When errors occur that prevent tape operation, they are reported persistently until
the problem is cleared. For medium-related errors (usually reported with a Sense
Key of 3), the error is reported until the cartridge is successfully unloaded. For
hardware-related errors (usually reported with a Sense Key of 4), the error is
reported until the drive successfully performs a power-on self test. These persistent
errors are only reported on those commands that are eligible for deferred Check
Condition reporting (see Table 4 on page 8). The error may or may not be reported
as Deferred.
Fencing Behavior
The Ultrium 2 and later drives will fence the drive (prevent certain operations)
when errors are detected that could endanger customer data if further usage is
allowed. The operations that are prevented depend on the nature of the error
encountered and the current drive state. The drive will post an FSC for the original
error that caused the fence condition. Then, fencing FSCs will be reported as status
to the attempted host commands that are not allowed due to fence.
The following Fence states are defined and used:

1. ALLOW_NO_OPERATION
a. All medium access commands (Read/Write/Motion) are rejected.
b. (SCSI/Panel/LDI) Unload is accepted.
c. After the cartridge is ejected:
v When load is attempted, the cartridge stays at mount position and Good
status is returned for TUR.
v From the above state, the cartridge can be ejected normally.
v Other medium access commands are rejected.
2. ALLOW_UNLOAD
c. Once a cartridge is ejected, Fence state is cleared. You can load a new
cartridge and perform all medium access commands.
3. ALLOW_ALLOCATE
c. Once a cartridge is ejected, Fence state is cleared. You can load a new
cartridge and perform all medium access commands.
d. Space command is rejected while in Fence state.
4. MID-TAPE RECOVERY
See the “Mode Page 2Fh: Behavior Configuration Mode Page” on page 134 for
a description of this behavior.
See Table 258 on page 263 for the errors that trigger the fence states.

Table 258. Fence State to Error Mapping
Description Fence State
v Severe Drive Hardware problem

v Severe Media Hardware problem
ALLOW_NO_OPERATION v Temperature Overrange
v Load or Unload Hardware problem
v Severe Firmware Problem
v Hardware Problem detected that could affect Writing

ALLOW_LOCATE
v Hardware Problem detected that could affect Reading
v Serious Drive Hardware problem — May be recovered

on a different mount
ALLOW_UNLOAD v Serious Media problem — Drive may be recovered on
different mount
v Serious Firmware problem — May be recovered on
different mount
MID-TAPE RECOVERY See “Mode Page 2Fh: Behavior Configuration Mode
Page” on page 134
Chapter 4. Error Sense Information 263

Chapter 5. Implementation Considerations
Device Clocks
The drive supports a Device Clock that maintains a timestamp for various items.
This timestamp gets recorded in drive error logs.
The timestamp origin (TORIG) is one of those specified in Table 259.

Table 259. Timestamp Origin
Value Definition
000b Timestamp initialized to zero at power-on
001b Reserved
010b Timestamp initialized by the SET TIMESTAMP command
011b Timestamp initialized by the Library over the Library Drive Interface (i.e. RS-422)
100b - 111b Reserved
Once a timestamp is initialized it begins counting from that time forward. Once
the timestamp is initialized it remains in effect until one of the following occurs:
v A SET TIMESTAMP command is processed;
v An LDI command is processed that modified the timestamp; or
v A Hard reset.
The method used is indicated in the extended control mode page.
The Timestamp is not affected by an I_T nexus loss or a Logical Unit reset.
The TIMESTAMP is specified in table Table 260.

Table 260. TIMESTAMP Format
Bit
Byte 7 6 5 4 3 2 1 0
0 (MSB)
TIMESTAMP
5 (LSB)
The TIMESTAMP field contains the value established at the last action that set the
timestamp incremented by one for every millisecond that has elapsed since the
timestamp was set.

Chapter 6. WORM BEHAVIORS
Conditions for Writing
If the following condition is met, writing is allowed:
v the cartridge is uninitialized
If all the following conditions are met, writing is allowed:

v the current logical position is at BOT
v there are only filemarks between here and EOD
If all of the following conditions are met, writing is allowed:

v if the current logical position is at BOT
v there are exactly 1 or 2 data records, followed by 0 to infinite number of
filemarks, followed by no data records, followed by EOD
If all of the following conditions are met, writing is allowed:

v the current logical position is BETWEEN BOT and EOD:
v there are only filemarks from the current logical position to EOD
v there is at least one filemark immediately before the current logical position
If the following condition is met, writing is allowed:

v the current logical position is AT® EOD
Command Behavior When WORM Medium Has Been Tampered With

Table 261 specifies the behavior of the drive when it has detected the WORM
medium that is loaded in the drive has been tampered with (See Sequential Access
Drive Configuration page).
Table 261. Behavior when the loaded medium has suspect integrity
Command WTRE=01b WTRE=00b or 10b
Write 0x7/300D 0x7/300D
Write Filemark n (n !=0) 0x7/300D 0x7/300D
Write Filemark 0 (buffered data) 0x7/300D 0x7/300D
Write Filemark 0 (no buffered data) Good Good
Erase 0x7/300D 0x7/300D
Read Good 0x3/300D
Verify Good 0x3/300D
Space Good 0x3/300D
Locate to (block !=0) Good 0x3/300D
Locate to 0 Good Good
Rewind Good Good
Unload Good Good
Load Good Good

Chapter 7. Sense Keys and Additional Sense
This chapter lists all possible combinations of Sense Keys, Additional Sense Codes
(ASC), and Additional Sense Code Qualifiers (ASCQ) that are reported by this
device.
Notes:
1. ASCs of X’EE’ and X’EF’ are used for encryption related features and are only
supported by encryption capable devices.
2. When using encryption enabled devices in an in-band configuration (System
method or key path), ASCs of X’EF’ are used to initiate a key management
session via a system proxy.
Sense Key 0 (No Sense)

Table 262. ASC and ASQ Summary for Sense Key 0 (No Sense)
ASC
ASCQ Description
00 00 No Additional Sense Information - (unsolicited, no CA/CC)
00 00 No Additional Sense Information - EOM=B’1’ (Early Warning)
00 00 No Additional Sense Information - ILI=B’1’
00 00 No Additional Sense Information - FM=B’1’
00 01 Filemark Detected
00 02 End-of-Partition/Medium Detected (Early Warning)
00 04 Beginning-of-Partition/Medium Detected
00 16 Operation in Progress
82 82 Drive Requires Clearing

Sense Key 1 (Recovered Error)
Table 263. ASC and ASQ Summary for Sense Key 1 (Recovered Error)
ASC
ASCQ Description
00 00 No Additional Sense Information
0C 00 Write Error: A write error occurred, but was recovered. Data was successfully written to tape.
11 00 Read Error: A read error occurred, but was recovered. Data was successfully read from tape.
17 01 Recovered Data with Retries
37 00 Mode Parameters Rounded
5D 00 Failure Prediction Threshold Exceeded
5D FF Failure Prediction Threshold Exceeded (FALSE)
EF 13 Encryption - Key Translate
Note: ASCs of EEh and EFh are used for encryption related features and are only supported by encryption capable
devices
Note: When using encryption enabled devices in an in-band configuration (System method or key path), ASCs of
X’EF’ are used to initiate a key management session via a system proxy.

Sense Key 2 (Not Ready)
Table 264. ASC and ASQ Summary for Sense Key 2 (Not Ready)
ASC
ASCQ Description
04 00 Logical Unit Not Ready, Cause Not Reportable: A tape is present in the drive, but it is in the
process of being unloaded.
04 01 Logical Unit Is in Process of Becoming Ready
04 02 Initializing Command Required: A tape is present in the drive, but it is not logically loaded.
04 12 Logical Unit Not Ready, Offline
30 03 Cleaning Cartridge Installed
30 07 Cleaning Failure
3A 00 Medium Not Present
3E 00 Logical Unit Has Not Self-configured
Sense Key 3 (Medium Error)

Table 265. ASC and ASQ Summary for Sense Key 3 (Medium Error)
ASC
ASCQ Description
04 10 Logical Unit Not Ready, Auxiliary Memory Not Accessible
09 00 Track Following Error (Servo)
0C 00 Write Error
11 00 Unrecovered Read Error
11 12 Auxiliary Memory Read Error
14 00 Recorded Entity Not Found
30 00 Incompatible Medium Installed
30 01 Cannot Read Medium, Unknown Format
30 02 Cannot Read Medium, Incompatible Format
30 0D Medium Error/WORM Medium - Integrity Check: Set when the drive rejects a Read operation
because the current cartridge is a Suspicious WORM cartridge, and the WTRE bit is set to 0.
31 00 Medium Format Corrupted
3B 00 Sequential Positioning Error
50 00 Write Append Error
51 00 Erase Failure
52 00 Cartridge Fault
53 00 Media Load or Eject Failed
53 04 Medium Thread or Unthread Failure
EE 60 Encryption - Proxy Command Error
EE D0 Encryption - Data Read Decryption Failure
EE D1 Encryption - Data Read after Write Decryption Failure
EE E0 Encryption - Key Translation Failure
Chapter 7. Sense Keys and Additional Sense 271

Table 265. ASC and ASQ Summary for Sense Key 3 (Medium Error) (continued)
ASC
ASCQ Description
EE E1 Encryption - Key Translation Ambiguous
EE F0 Encryption - Decryption Fenced (Read)
EE F1 Encryption - Encryption Fenced (Write)
devices.
EFh are used to initiate a key management session via a system proxy.

Sense Key 4 (Hardware or Firmware Error)
Table 266. ASC and ASQ Summary for Sense Key 4 (Hardware or Firmware Error)
ASC
ASCQ Description
04 03 Manual Intervention Required: A tape is present in the drive but could not be loaded or unloaded
without manual intervention.
08 01 Logical Unit Communication Failure
40 XX Diagnostic Failure: The Additional Sense Code Qualifier indicates the failing component.
41 00 Data Path Failure
44 00 Internal Target Failure
51 00 Erase Failure
53 00 Media Load or Eject Failed
53 04 Medium Thread or Unthread Failure
EE 0E Encryption - Key Service Timeout
EE 0F Encryption - Key Service Failure
devices.
EFh are used to initiate a key management session via a system proxy.
Sense Key 5 (Illegal Request)

Table 267. ASC and ASQ Summary for Sense Key 5 (Illegal Request)
ASC
ASCQ Description
1A 00 Parameter List Length Error
20 00 Invalid Command Operation Code
24 00 Invalid Field in CDB
25 00 Logical Unit Not Supported
26 00 Invalid Field in Parameter List
26 11 Encryption - Incomplete Key-Associate Data Set
29 04 Device Internal Reset
2C 00 Command Sequence Error
2C 0B Not Reserved - The OIR bit of the Sequential Access Device page is set and the I_T nexus
attempting to communicate with the drive does not hold a reservation
3B 0C Position Past Beginning of Medium: A command that required the medium to be at BOP was
attempted when the medium was not at BOP (for example, Set Capacity).
49 00 INVALID MESSAGE ERROR
53 02 Medium Removal Prevented
82 83 Bad Microcode Detected: The data transferred to the drive during a firmware upgrade is
corrupted or incompatible with the drive hardware.
A3 01 OEM Vendor-Specific
EE 00 Encryption - Key Service Not Enabled

Table 267. ASC and ASQ Summary for Sense Key 5 (Illegal Request) (continued)
ASC
ASCQ Description
EE 01 Encryption - Key Service Not Configured
EE 02 Encryption - Key Service Not Available
EE 10 Encryption - Key Required
EE 20 Encryption - Key Count Exceeded
EE 21 Encryption - Key Alias Exceeded
EE 22 Encryption - Key Reserved
EE 23 Encryption - Key Conflict
EE 24 Encryption - Key Method Change
EE 25 Encryption - Key Format Not Supported
EE 26 Encryption - Unauthorized Request - dAK
EE 27 Encryption - Unauthorized Request - dSK
EE 28 Encryption - Unauthorized Request - eAK
EE 29 Encryption - Authentication Failure
EE 2A Encryption - Invalid RDKi
EE 2B Encryption - Key Incorrect
EE 2C Encryption - Key Wrapping Failure
EE 2D Encryption - Sequencing Failure
EE 2E Encryption - Unsupported Type
EE 2F Encryption - New Key Encrypted Write Pending
EE 30 Encryption - Prohibited Request
EE 2D Encryption - Sequencing Failure
EE 30 Encryption - Prohibited Request
EE 31 Encryption - Key Unknown
EE 42 Encryption - EKM Challenge Pending
EE E2 Encryption - Key Translation Disallowed
EE FF Encryption - Security Prohibited Function
EF 01 Encryption - Key Service Not Configured
devices.
X’EF’ are used to initiate a key management session via a system proxy.

Sense Key 6 (Unit Attention)
Table 268. ASC and ASQ Summary for Sense Key 6 (Unit Attention)
ASC
ASCQ Description
28 00 Not Ready to Ready Transition, Medium May Have Changed
28 01 Import or Export Element Accessed
29 00 Power On, Reset, or Bus Device Reset Occurred: This also occurs on Fibre Channel drives when a
host logs in (PLOGI).
29 05 Transceiver Mode Changed To Single-ended: The drive detected a transceiver mode change from
LVD mode to MSE mode
29 06 Transceiver Mode Changed To LVD: The drive detected a transceiver mode change from MSE
mode to LVD mode
2A 01 Mode Parameters Changed
2A 11 Encryption - Data Encryption Parameters Changed by Another I_T Nexus
2A 12 Encryption - Data Encryption Parameters Changed by Vendor Specific Event
2F 00 Commands Cleared by Another Initiator
3F 03 Inquiry Data Has Changed
3F 0E Reported LUNs Data Has Changed
5D FF Failure Prediction False: A MODE SELECT command has been used to test the Failure Prediction
System.
5A 01 Operator Medium Removal Request
EE 12 Encryption - Key Change Detected
EE 18 Encryption - Changed (Read)
EE 19 Encryption - Changed (Write)
EE 40 Encryption - EKM Identifier Changed
EE 41 Encryption - EKM Challenge Changed
EE 50 Encryption - Initiator Identifier Changed
EE 51 Encryption - Initiator Response Changed
devices.
devices.
Sense Key 7 (Data Protect)

Table 269. ASC and ASQ Summary for Sense Key 7 (Data Protect)
ASC
ASCQ Description
26 10 Encryption - Data Decryption Key Fail Limit
27 00 Write Protected
2A 13 Encryption - Data Encryption Key Instance Counter Has Changed
30 05 Cannot Write Medium, Incompatible Format

Table 269. ASC and ASQ Summary for Sense Key 7 (Data Protect) (continued)
ASC
ASCQ Description
30 0D Data Protect/WORM Medium - Integrity Check: Set when the drive rejects a Write operation
because the current cartridge is a Suspicious WORM cartridge.
30 0C Data Protect/WORM Medium - Overwrite Attempted: Set when the drive rejects a Write
operation because the rules for allowing WORM writes have not been met.
52 00 Cartridge Fault
74 00 Security Error
74 01 Encryption - Unable to Decrypt Data
74 02 Encryption - Unencrypted Data Encountered While Decrypting
74 03 Encryption - Incorrect Data Encryption Key
74 04 Encryption - Cryptographic Integrity Validation Failed
74 05 Encryption - Error Decrypting Data
EF 10 Encryption - Key Required
EF 11 Encryption - Key Generation
EF 13 Encryption - Key Translate
EF 1A Encryption - Key Optional
EF C0 Encryption - No Operation
devices.
devices.
Sense Key 8 (Blank Check)

Table 270. ASC and ASQ Summary for Sense Key 8 (Blank Check)
ASC
ASCQ Description
00 05 End-of-Data (EOD) Detected
0B 01 The drive detected an overtemperature condition. The currently loaded medium has been ejected.
14 03 End-of-Data (EOD) not Found

Sense Key B (Aborted Command)
Table 271. ASC and ASQ Summary for Sense Key B (Aborted Command)
ASC
ASCQ Description
00 02 End-of-Partition/Medium Detected
0E 01 INFORMATION UNIT TOO SHORT
3D 00 Invalid Bits in Identify Message
3F 0F Echo Buffer Overwritten
43 00 Message Error
45 00 Select/Reselect Failure
48 00 Initiator Detected Error Message Received
49 00 Invalid Message Error
4A 00 Command Phase Error
4B 00 Data Phase Error
4B 02 TOO MUCH WRITE DATA
4B 03 ACK/NAK TIMEOUT
4B 04 NAK RECEIVED
4B 05 DATA OFFSET ERROR
4B 06 INITIATOR RESPONSE TIMEOUT
4E 00 Overlapped Commands
Sense Key D (Volume Overflow)

Table 272. ASC and ASQ Summary for Sense Key D (Volume Overflow)
ASC
ASCQ Description
00 02 End-of-Partition/Medium Detected

Chapter 8. Device Hardware Encryption (Ultrium 4)
Ultrium 4 devices contain hardware which performs user data write encryption
and read decryption, protecting all user data written to the medium from
unauthorized use [provided it is integrated into a secure system design].

Chapter 9. Attachment Features
Types of Interface Attachments
The Ultrium Tape Drive communicates with servers that use SCSI parallel, Fibre
Channel, or SAS interfaces. The interfaces share certain tape LUN behaviors, but
also possess unique features. This chapter describes the common and unique
features of these interfaces.
Common Tape LUN Behaviors

Parallel SCSI and Fibre Channel and SAS interfaces share the following tape LUN
behaviors:
v Power-on procedure
v Reset strategy
v Abort handling
v Multi-initiator support
v Status codes
The sections that follow describe each behavior.
Power-On
The drive responds to INQUIRY, REPORT LUNs, REQUEST SENSE, and TEST
UNIT READY commands within 5 seconds of power-on. The first command (other
than INQUIRY or REQUEST SENSE) from any initiator gets a Check Condition
status with Unit Attention sense data for the power-on. After this, any medium
access command is reported with a Sense Key of Not Ready and an ASC/ASCQ of
LUN Has Not Self-Configured Yet (3E00). Once the drive has completed its self test
and setup procedures, the drive attempts to load any tape that is present. Medium
access commands are reported with an ASC/ASCQ of Drive in Process of
Becoming Ready (0401).
Reset Strategy
The drive supports the hard reset option as is required by SCSI-3. On receiving a
reset, the following actions are taken:
v The current I/O process is aborted, as in “Abort Handling” on page 282.
v Any queued I/O processes from other initiators are removed.
v All reservations are cleared.
v All mode values are reset to their defaults.
v Synchronous/Wide negotiations are cleared (applies only to SCSI parallel
attach).
v A unit attention condition is set.
v A logical position is established that may or may not be the same as the position
prior to the reset. Where possible, the logical position prior to reset is
maintained.
For drives that use he Fibre Channel or SAS interface, the next command that is
eligible for the Unit Attention Check Condition from each initiator gets a Check

Condition status, with Unit Attention sense data for the reset. However, other
commands may not be processed until the internal state of the drive has been
reset.
Drives that use a SCSI interface are able to respond to the INQUIRY, REPORT
LUNs, REQUEST SENSE, and TEST UNIT READY commands within 250 ms of the
reset line being released. The next command that is eligible for the Unit Attention
Check Condition from each initiator gets a Check Condition status, with Unit
Attention sense data for the reset. However, other commands may not be
processed until the internal state of the drive has been reset. Any commands that
allow Disconnects and cannot be processed are accepted and queued until the
drive is ready to process them. Those commands that do not allow Disconnects
receive Busy status response.
Abort Handling
If a command is aborted on a drive with a Fibre Channel or SAS interface, see
Table 273 for abort processing.
If a command is aborted on a drive with a SCSI interface, one of the following

conditions will occur:
v If an abort condition is detected before a command phase completes, the bus is
set to Bus Free and the command is not executed.
v If an abort condition is detected during status phase, the bus is set to bus free.
v If an abort condition is detected between the end of the command phase and the
start of the status phase, then the bus is set to Bus Free and the processing in
Table 273 on page 282 is carried out.
Table 273. Abort Condition Handling
Command Abort Processing
ERASE Long erase is aborted as quickly as possible without corrupting tape format.
Short erase completes.
INQUIRY None.
LOAD/UNLOAD Load completes and logically positions tape at BOM. Unload is aborted,
leaving logical position at BOM unless operation is past the ‘point of no
return’, in which case the tape is ejected.
LOCATE The logical position is set back to that at the start of the operation unless the
operation is past its ‘point of no return’, in which case the operation
completes.
LOG SELECT If data transfer is completed, command is completed; otherwise, no action is
taken.
LOG SENSE None.
MODE SELECT If data transfer is completed, command is completed; otherwise, no action is
taken.
MODE SENSE None.
PERSISTENT RESERVE IN None.
PERSISTENT RESERVE OUT If data transfer is completed, the command is completed; otherwise, no
action is taken.
PREVENT/ALLOW MEDIUM The command completes.
REMOVAL
READ The current position is set to the first record boundary at or after the start of
the current data burst.

Table 273. Abort Condition Handling (continued)
Command Abort Processing
READ ATTRIBUTE None.
READ BLOCK LIMITS None.
READ BUFFER None.
READ POSITION None.
RECEIVE DIAGNOSTIC RESULTS None.
RELEASE UNIT The command completes.
REPORT DENSITY SUPPORT None.
REPORT LUNs None.
REQUEST SENSE Sense data is discarded.
RESERVE UNIT The command completes.
REWIND The command completes.
SEND DIAGNOSTIC Vendor unique.
SPACE The logical position is set back to that at the start of the operation unless the
operation is past its ‘point of no return’, in which case the operation
completes.
TEST UNIT READY None.
VERIFY The logical position is set to the next record boundary after the point where
the verify was aborted.
WRITE The data up to first record boundary in the current burst is written to buffer
or tape, depending on Buffered Mode. Any subsequent data is discarded. If
there is no record boundary in the current burst, the record is truncated to
the amount of data transferred and written to buffer or tape, again
depending on Buffered Mode.
WRITE BUFFER If data transfer is completed, the command is completed; otherwise, no
action is taken.
WRITE FILE MARKS The command completes.
For drives with a SCSI interface, if a command other than INQUIRY, REPORT
LUNs, REQUEST SENSE, or TEST UNIT READY is received after the abort but
before the drive is ready to process the command, the drive attempts to disconnect
and wait until the abort processing has completed before executing the command.
If disconnects are not allowed, Busy status is returned. A TEST UNIT READY
command reports with status immediately.
For drives with a Fibre Channel or SAS interface, an INQUIRY command returns
the required data and gives Good status. A REQUEST SENSE command gives no
sense. A TEST UNIT READY command reports with status immediately.
Chapter 9. Attachment Features 283

Multi-initiator Support
SCSI-attached drives support a maximum of two initiators on the same bus.
Fibre-Channel-attached drives and Serial Attached SCSI drives support an infinite
number of initiators, but have a limit on how many initiators can be logged in
processing commands concurrently. When this limit is exceeded, the least recently
used (LRU) initiator that is not reserved or does not have an outstanding
command will be implicitly logged out.
The drive supports untagged queuing when operating with multiple initiators. If a
command from one initiator is being processed when a command other than
INQUIRY, REPORT LUNs, REQUEST SENSE, and TEST UNIT READY is received
from a second initiator, the new command is queued. Media access commands (for
example, Write, Write Filemarks, Read, Verify, Rewind, Mode Select that changes
block size) are always executed in strict order of receipt. For drives with a SCSI
interface, if the queue is full or disconnect privilege is not granted in the new
command, the drive reports busy status.
The INQUIRY, REPORT LUNs, REQUEST SENSE, and TEST UNIT READY
commands are always processed immediately, irrespective of whether a command
from another initiator is being processed.
The drive maintains sense data for the supported number of initiators. On
Fibre-Channel-attached drives and Serial Attached SCSI drives, if an additional
initiator connects to the drive and causes an initiator to be implicitly logged out,
the drive erases all sense data for that initiator before processing the command for
the new initiator. See “Sense Data Management” on page 261 for more details of
sense data management.
Status Codes
Table 274. Status Codes
Status Code Value Circumstance
Good 00h The command completed without problems.
Check Condition 02h A problem occurred during command execution. The sense data should be
examined to determine the nature of the problem.
Busy 08h The drive is unable to accept the command at this time. This status is
returned during the power-on sequence or if there are commands from too
many initiators outstanding. (See “Multi-initiator Support” on page 284.) It
is also returned when commands are issued without Disconnect Privilege
and when another command is in progress.
Reservation Conflict 18h This status is returned if the drive is reserved for an initiator other than
the one sending the command.
Queue Full 28h Not normally returned.

Features of the SCSI Interface
The Ultrium Tape Drive’s SCSI parallel interface features the following:
v LUN identification
v Bus parity errors
v Disconnect strategy
v Messages
LUN Identification
Identify messages are used to identify the LUN to which an initiator is connecting
and to identify which LUN is reconnecting to an initiator. These are required. The
LUN field in SCSI-2 commands is not used.
SPI Information Unit (Ultra 320 Attached Drives Only)

The following TASK MANAGEMENT FUNCTIONs are supported in SPI
Command Information Unit:
01h - ABORT TASK (see “Abort Handling” on page 282)

02h - ABORT TASK SET (see “Abort Handling” on page 282)
04h - CLEAR TASK SET (see “Abort Handling” on page 282)
08h - LOGICAL UNIT RESET (see “Abort Handling” on page 282)
20h - TARGET RESET (see “Abort Handling” on page 282)
The following TASK ATTRIBUTE CODEs are supported in SPI Command

Information Unit:
20h - SIMPLE
Bus Parity Errors (When Using Data Group Transfers)

On detecting a bus parity error during a Command or Data Out phase or receiving
an Initiator Detected Error message during a Data In or Status phase, the drive
attempts to retry the Bus phase. A Restore Pointers message is sent to the initiator
and the transfer is repeated. Only one retry is attempted for any given burst. If the
retry fails or the Restore Pointers message is rejected by an Initiator Detected Error,
Message Reject, or Message Parity message, then the drive goes to the Status phase
and attempts to report Check Condition status. If this fails with an Initiator
Detected Error message, the drive goes to Bus Free. The Sense Key is set to
Aborted Command and the ASC/ASCQ is set to Command Phase Error (4A00) (if
the error was in the Command phase) or to Data Phase Error (4B00) (if the error
was in the Data phase). If the error was in the Status phase, the sense data remains
as that from the command.
If an Initiator Detected Error or Message Parity Error message is received during

the Message In phase, the initiator has detected an error in the message. The drive
goes to Message In and resends the message that was in error. If the subsequent
message is rejected with an Initiator Detected Error, then the drive goes to the
Status phase and sends Check Condition status. The Sense Key is set to Aborted
Command and the ASC/ASCQ is set to Initiator Detected Error (4800). If the
subsequent message is rejected with a Message Parity Error, then the drive goes to
the Status phase and sends Check Condition status. The Sense Key is set to
Aborted Command and the ASC/ASCQ is set to Message Error (4300).

On detecting a bus parity error during a Message Out phase, the drive handshakes
in all message bytes until ATN is deserted. It then stays in the Message Out phase
to indicate to the initiator that the whole Message Out phase should be resent.
Bus Parity Errors (When Using Information Unit Transfers)

On detecting an iuCRC error in a SPI L_Q information unit, the drive goes bus
free. If the retry fails, then the drive goes to the SPI Status information unit and
attempts to report Check Condition status. If this fails with an Initiator Detected
Error message, the drive goes Bus Free. The Sense Key is set to Aborted Command
and the ASC/ASCQ is set to INFORMATION UNIT iuCRC ERROR DETECTED
error (4703h) (if the error was in the SPI Command information unit, SPI Data
information unit or Data Stream information unit). If the error was in the Status
information unit, the sense data from the command remains.
Disconnect Strategy
The disconnect strategy used by the drive is based on the assumption of
maximizing bus utilization for large sequential data transfers from a large data
buffer. The drive disconnects whenever it believes that it can provide better bus
utilization. This may be between Command and Data phases, between bursts of
data, or before sending status. However, the drive guarantees that it sends the
configured maximum burst size or the remaining data in the transfer in any single
Data phase burst if the maximum burst size has been set to a value other than 0.

Messages
Message Out
Table 275. Supported Outbound Messages
Support1 Name Code Support
An abort condition is generated (see “Abort Handling” on
D Abort 06h
page 282).
A reset condition is generated (see “Abort Handling” on
D Bus Device Reset 0Ch
page 282).
E Abort Test (Tag) 0Eh
page 282
E Clear Task Set 0Eh
page 282 for each task in the task set.
The task is to use Simple tagged queueing using the
E Simple (queue tag) 20h/Tag(00h - FFh)
queue tag as a qualifier on the task.
The Identify Out message is sent by the initiator to
identify the Logical Unit to be accessed and to set
Disconnect Privilege. The LUNTAR flag must be 0. The
D Identify 80h+
Identify Out message must be sent as the first thing after
selection. If it is sent at any other time, the drive responds
with a Message Reject message and goes to Bus Free.
The initiator has detected an error in the data being sent
in a Message Command, Data, or Status phase. The drive
retries the data burst or message (see “Bus Parity Errors
Initiator Detected
D, E 05h (When Using Data Group Transfers)” on page 285). If the
Error
message is received immediately after an Identify
message or after the Command Complete message has
been sent, the drive goes to Bus Free.
The initiator has detected a parity error in a message. The
drive retries the message (see “Bus Parity Errors (When
Using Data Group Transfers)” on page 285). If the
Message Parity
D, E 09h message is received during a Command, Data, or Status
Error
phase, immediately after an Identify message, or after the
Command Complete message has been sent, the drive
goes to Bus Free.
This message is sent when the initiator does not support
a message sent by the drive or considers the message
inappropriate. If the message being rejected is Disconnect,
Synchronous Data Transfer Request, or Wide Data
Transfer Request, the operation continues without those
D, E Message Reject 07h features. For all other messages except Restore Pointers,
the message is treated as an Abort message. If the
message is received during a Command, Data, or Status
phase, immediately after an Identify message, or after the
Command Complete message has been sent, the drive
goes to Bus Free.
D, E No Operation 08h This message has no effect and is ignored.

Table 275. Supported Outbound Messages (continued)
transfer negotiation. The initiator is expected to do so.
Since this mode is most important at power-on, no SCSI
method of changing this mode is provided. Instead, an
internal method of changing this mode is provided (VPD
or jumper setting). If the message is received after
Wide Data
D, E 01h,02h,03h selection and before the Command phase, it then goes to
Transfer Request
the Message In phase and responds with a valid response
to complete the negotiation. If the message is received at
any other time, a Message Reject is sent in response,
placing the Bus Width to 1 byte wide and data transfer
mode to asynchronous for that initiator.
The default mode is for the drive to never initiate a
Synchronous data transfer negotiation. The initiator is
expected to do so. Since this mode is most important at
power-on, no SCSI method of changing this mode is
provided. Instead, an internal method of changing this
Synchronous Data mode is provided (VPD or jumper setting). If the message
D, E 01h,03h,01h
Transfer Request is received after selection and before the Command
phase, it then goes to the Message In phase and responds
with a valid response to complete the negotiation. If the
message is received at any other time, a Message Reject is
sent in response, placing the drive data transfer mode for
that initiator to asynchronous
Parallel Protocol Request messages are used to negotiate a
Parallel Protocol synchronous transfer agreement, a wide data transfer
D,E 01h,06h,04h
Request agreement, and set the protocol options between the
initiator and the target.
1
D - IU Transfers Disables
E - IU Transfers Enabled

Message In
Table 276. Supported Inbound Messages
This message is sent by the drive at the end
Command
of the Status phase to indicate that a
Complete
D 00h command is complete. Once the message is
(Task
sent, the drive releases the bus and goes to
Complete)
Bus Free.
This message is sent by the drive to indicate
that it is about to disconnect from the bus
and go to Bus Free. During a Data phase, it
D Disconnect 04h is always preceded by a Save Data Pointers
message. If a Message Reject message is
received in response to this message, then
the disconnect is prevented.
The Identify In message is sent to the
initiator during reconnect to indicate which
D Identify 80h+
Logical Unit is reconnecting. The Disconnect
Privilege and LUNTAR flags are both clear.
This message is sent by the drive to the
Ignore Wide initiator to indicate that a byte on a wide
D 23h
Residue bus is not valid. This is supported whenever
a wide transfer is active.
This message is sent when the initiator does
not support a message sent by the drive or
considers the message inappropriate. If the
message being rejected is Disconnect,
Synchronous Data Transfer Request, or Wide
Data Transfer Request, the operation
Message continues without those features. For all
D, E 07h
Reject other messages except Restore Pointers, the
message is treated as an Abort message. If
the message is received during a Command,
Data, or Status phase, immediately after an
Identify message, or after the Command
Complete message has been sent, the drive
goes to Bus Free.
Parallel Protocol Request messages are used
Parallel to negotiate a synchronous transfer
D, E Protocol 01h,06h,04h agreement, a wide data transfer agreement,
Request and set the protocol options between the
initiator and the target.
This message causes the initiator to reset its
data transfer pointers to the values they held
when the last Save Data Pointers message
Restore
D 03h was sent. It is sent when a parity error is
Pointers
detected on the bus or when an Initiator
Detected Error message is received in order
to retry the Data phase.
This message instructs the initiator to save
its current data transfer pointers for use with
Save Data
D 02h a subsequent Restore pointers message. This
Pointers
message is always sent before a Disconnect
message during Data phases.

Table 276. Supported Inbound Messages (continued)
The default mode is for the drive to never
initiate a Synchronous data transfer
negotiation. The initiator is expected to do
so. Since this mode is most important at
power-on, no SCSI method of changing this
mode is provided. Instead, an internal
method of changing this mode is provided
Synchronous
(VPD or jumper setting). If the message is
D, E Data Transfer 01h,03h,01h
received after selection and before the
Request
Command phase, it then goes to the
Message In phase and responds with a valid
response to complete the negotiation. If the
message is received at any other time, a
Message Reject is sent in response, placing
the drive data transfer mode for that
initiator to asynchronous
The default mode is for the drive to never
initiate a Wide data transfer negotiation. The
initiator is expected to do so. Since this
mode is most important at power-on, no
SCSI method of changing this mode is
provided. Instead, an internal method of
changing this mode is provided (VPD or
Wide Data
jumper setting). If the message is received
D, E Transfer 01h,02h,03h
after selection and before the Command
Request
phase, it then goes to the Message In phase
and responds with a valid response to
complete the negotiation. If the message is
received at any other time, a Message Reject
is sent in response, placing the Bus Width to
1 byte wide and data transfer mode to
asynchronous for that initiator.
1
D - IU Transfers Disables
E - IU Transfers Enabled

Features of the Fibre Channel Interface
The Ultrium Internal Tape Drive Model T200F (the Fibre Channel Generation 1
drive) is an Arbitrated-Loop-only device (also known as an NL port). The drive
supports Fibre Channel Arbitrated Loop (FC-AL) protocol, and uses Class 3 Service
frames. The drive also supports both public (switch-attached) and private loops.
Like the Ultrium Internal Tape Drive, Ultrium 2 and later drives can attach as an
FC-AL device (that is, L-Port). The drives also support operating as a public
(switch-attached) or private device (that is, L-Port to FL-Port; or L-Port to L-Port).
Ultrium 2 and later drives can also attach using the point-to-point protocol (also
known as an N-Port). When operating in the point-to-point protocol, the Ultrium 2
and later drives can attach in a Fabric topology (that is, N-Port to F-Port).
The World Wide Node Name and Port Name that are used by an Ultrium Tape
Drive follow the format of the Institute of Electrical and Electronics Engineers
(IEEE).
The IBM Ultrium Tape Drive is compliant with the FC-Tape Technical Report of the
Accredited Standard Committee NCITS. IBM recommends that your server’s device
driver and host bus adapter (HBA) use the Class 3 Error Recovery procedures that
are specified in the Fibre Channel Protocol for SCSI, Second Version (FCP-2).
Features of the Serial Attached SCSI (SAS) Interface

The World Wide Node Name and Port Name that are used by an Ultrium Tape
Drive follow the format of the Institute of Electrical and Electronics Engineers
(IEEE).
The IBM Ultrium Tape Drive is compliant with the American National Standard,
Project T10/1601-D, Information technology - Serial Attached SCSI - 1.1 (SAS-1.1),
Revision 10, 21 September 2005.

Chapter 10. Firmware Download
This chapter describes Firmware download.
Identifying Level Hardware of Drive (Not Ultrium 1)

The firmware that is loaded in the drive will report a Load ID and RU Name in
Inquiry page 03h. The Load ID and RU Name are used to designate the Level
Hardware. The following table defines the Load ID and RU Name values for each
Level Hardware.

Table 277. Load ID and RU Name Designation
RU Name (EBCDIC F/W Product ID
Product (Level string) (Hex (Min (Left-Aligned) Product ID (Left-Aligned)
Hardware) Load ID representation) Level) (IBM) (OEM)
Ultrium 2 Ultra ″AJEFGP99″
A1700299 -
160 0xC1D1C5C6C7D7F9F9
″AJEFGP9A″
Ultrium 2 FC 2G A170029A -
0xC1D1C5C6C7D7F9C1
ULT3580-TD2 ULTRIUM-TD2
Ultrium 2 Ultra ″AJEFAX67″
A1700D67 -
160 RoHS 0xC1D1C5C6C1E7F6F7
Ultrium 2 FC 2G ″AJEFAX68″
A1700D68 -
RoHS 0xC1D1C5C6C1E7F6F8
Ultrium 3
Ultrium 3 Ultra
-
160 ″AJEFAX50″
A1700D50
Ultrium 3 Ultra 0xC1D1C5C6C1E7F5F0
5BG2
160 RoHS ULT3580-TD3 ULTRIUM-TD3
Ultrium 3 FC2 -
″AJEFAX52″
Ultrium 3 FC4 A1700D52 5BG2
0xC1D1C5C6C1E7F5F2
RoHS
Ultrium 3 A1700D64 ″AJEFAX64″
Half-High SAS 0xC1D1C5C6C1E7F6F4
A1700D53 ″AJEFAX53″
0xC1D1C5C6C1E7F5F3 - ULT3580-HH3 ULTRIUM-HH3
Ultrium 3 A1700D63 AJEFAX69
Half-High Ultra
A1700D69 ″AJEFAX69″
160
0xC1D1C5C6C1E7F6F9
Ultrium 4
Ultrium 4 U160 A1700D6A ″AJEFAX6A″
0xC1D1C5C6C1E7F6C1
Ultirum 4 FC4 A1700D6B ″AJEFAX6B″
0xC1D1C5C6C1E7F6C2
- ULT3580-TD4 ULTRIUM-TD4
Ultrium 4 SAS A1700D6C ″AJEFAX6C″
0xC1D1C5C6C1E7F6C3
Ultrium 4 U320 A1700D70 ″AJEFAX70″
0xC1D1C5C6C1E7F7F0
- Not Applicable

Identifying the Level Hardware for Which the Firmware Image is
Intended (Not Ultrium 1)
The Firmware Image is defined in Table 278.
Table 278. Firmware Image
Bit Byte 7 6 5 4 3 2 1 0
0 MSB Not Specified
3 LSB
4 MSB Firmware Length + Header Length (m)
7 LSB
8 Load ID
(See Inquiry Page 03h)
11
12 Firmware Revision Level
(See Standard Inquiry bytes 32 - 35)
15
16 Reserved
23
24 RU Name
(See Inquiry Page 03h)
31
32 Not Specified
m
The Load ID and RU Name fields in the Firmware Image are used to define the
Level Hardware for which the Firmware Image is intended.
Download Process
1. Check the Level Hardware of the drive whose firmware is to be updated. See
“Firmware Designation Page (03h)” on page 28.
2. Confirm the Level Hardware of the Firmware Image to be loaded matches the
Level Hardware of the drive. (See Table 278.)
3. Download the Firmware Image. See “WRITE BUFFER” on page 258.
Chapter 10. Firmware Download 295

Appendix A. ASN.1 Information (Used for Encryption)
ASN.1 [DER] Encoding Implementation
See the appendix for an overview on ASN.1 with DER encoding. Within this
construct there are certain fields and layouts which are expected by the device.
These are detailed as follows:
SEDK ASN.1 Format

The SEDK structure must be understood by both the tape device and the EKM and
is the method for transferring the DK from the EKM to the drive.
MediaEncryptionStandardData ::= SEQUENCE {
version INTEGER, -- Always set to 0
mesInfo TbsMesInfo, -- Signed portion
mesSigAlgIDAlgorithmIdentifier OPTIONAL,-- No sig if clear key
OID= 1 3 14 3 2 11RSA signature
OID= 1 3 14 3 2 29RSA w/SHA1 signature
OID= 1 2 840 113549 1 1 5sha1withRSAEncryption
mesSignatureOCTET STRING OPTIONAL
TbsMesInfo ::= SEQUENCE {

driveDK [2] MesKeyInfo OPTIONAL-- Key blob for drive
MesKeyInfo ::= SEQUENCE {

xedkDataMesKeyData -- actual key information
MesKeyData ::= CHOICE {

kedkInfo[0] KeyEncryptedDKInfo, -- KEK/dSK/dAK encrypted DK
NOTE: SEDK(s) cannot use the PasswordEncryptedDKInfo (drive does not understand)
cdkInfo [2] ClearDK -- Clear DK (driveDK [SEDK ONLY])
redkInfo[3] KeyEncryptedDKInfo, -- KEK/dSK/dSK encrypted raw key
(does not use ClearKey)
NOTE: The "redkInfo" format does not contain any DK meta information [this could be changed
by using a different structure than KeyEncryptedDKInfo (or altered format)] and is
intended for environments where the DK generation and wrapping are performed entirely in
hardware. In some instances the key size may be unusual (i.e., a 192 bit odd parity 3DES
key [168 bits of entropy]) which the drive must expand to 256 in a common manner for use
as an AES key.
KeyEncryptedDKInfo ::= SEQUENCE {

kekid KEKeyIdentifier, -- Identifies KEK used by EKM
kekAlg AlgorithmIdentifier,
OID= 1 2 840 113549 1 1 1RSA
OID= 1 2 840 113549 1 1 7RSAES-OAEP
OID= TBD (ANY (for EEDK), used by EKM only)
kedk OCTET STRING, -- Encrypted ClearDK (if kedkInfo)
-- Encrypted raw key (if redkInfo)
dkUUID OCTET STRING OPTIONAL -- DKi

}
KEKeyIdentifier ::= CHOICE {

kekPublicDigest[0] DigestInfo-- Digest of public key,
OID= 1 3 14 3 2 26SHA-1
OID= 1 2 840 113549 2 7Legacy format
kekIssuerSNDigest[1] DigestInfo, -- Digest of DER encoded IssuerAndSerialNumber
kekLabel [2] IA5String(1::64)
ClearDK ::= SEQUENCE {

dataKeyAlg AlgorithmIdentifier, -- Algorithm used for data encryption
OID= 2 16 840 1 101 3 4 1AES (TBD 256-GCM specific?)
dataKey OCTET STRING -- The DK
dkUUID OCTET STRING OPTIONAL -- DKi
ASN.1 Standalone SEDK (SEDK Only, Clear, Signed) - Example

ASN1 Data
Ofs Ofs Len: ASN.1 Field Name
0 4 356: SEQUENCE {
: <30 82 01 64>
4 6 1 : INTEGER 0 version
: <02 01 00>
7 9 76 : SEQUENCE {mesInfo:
<30 4B>
: <02 01 00>
12 14 53 : [2] SEQUENCE { driveDK:
<A2 35>
14 16 69 : [2] SEQUENCE { cdkInfo
: <A2 45>
: <02 01 00>
19 21 12 : SEQUENCE { dataKeyAlg
: <30 0C>
21 23 8 : OID 2 16 840 1 101 3 4 1:
<06 08 60 86 48 01 65 03 04 01>
31 33 0 : NULL
: <05 00>
: }
33 35 32 : OCTET STRING dataKey:
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
67 69 16 : OCTET STRING dkUUID (OPTIONAL)
: <04 10>
: }
: }
: }
85 87 13 : SEQUENCE {mesSigAlgID
: <30 0D>
87 89 9 : OID 1 2 840 113549 1 1 5
: <06 09 2A 86 48 86 F7 0D 01 01 05>
98 100 0 : NULL
: <05 00>
: }
100 104 256: OCTET STRING mesSignature
: <04 82 01 00>
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??

: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: }
ASN.1 Standalone SEDK (SEDK Only, dSK Encrypted, Signed) -

Example
ASN1 Data
0 4 627 : SEQUENCE {
: <30 82 02 73>
4 6 1 : INTEGER version
: <02 01 00>
7 11 345 : SEQUENCE {mesInfo
: <30 82 01 59>
: <02 01 00>
14 18 338 : [2] SEQUENCE {driveDK
: <A2 82 01 52>
18 22 334 : [0] SEQUENCE {kedkInfo
: <A0 82 01 4E>
: <02 01 00>
25 27 33 : [0] SEQUENCE { kekid:
<A0 21> (kekPublicDigest)
27 29 31 : SEQUENCE { digestAlgorithm
: <30 1F>
29 31 5 : OID 1 3 14 3 2 26
: <05 2B 0E 03 02 1A>
36 38 0 : NULL
: <05 00>
38 40 20 : OCTET STRING digest (of dSK)
: <04 14>
: <?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ??>
: }
: }
60 62 13 : SEQUENCE { kekAlg
: <30 0D>
62 64 9 : OID 1 2 840 113549 1 1 1
: <06 09 2A 86 48 86 F7 0D 01 01 01>
73 75 0 : NULL
: <05 00>
: }
75 79 256 : OCTET STRING kedk
: <04 82 01 00>
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
Appendix A. ASN.1 Information (Used for Encryption) 299

: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
335 337 16 : OCTET STRING dkUUID (OPTIONAL)
: <04 10>
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: }
: }
: }
353 355 13 : SEQUENCE {mesSigAlgID
: <30 0D>
355 357 9 : OID 1 2 840 113549 1 1 5
: <06 09 2A 86 48 86 F7 0D 01 01 05>
366 368 0 : NULL
: <05 00>
: }
368 372 256 : OCTET STRING mesSignature
: <04 82 01 00>
: < ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: }
xAK/xSK ASN.1 Format

The dAK/eAK/dSK are public keys which are sent between the tape drive and
EKM in various messages. The dAK and eAK do not need to be protected in any
fashion. The dSK is pseudo secret and is used to protect the DK in the SEDK - this
effectively provides another layer of authentication.
KeyExchangeData ::= SEQUENCE {
kedInfo SubjectPublicKeyInfo, -- Signed portion
kedSigAlgID AlgorithmIdentifier OPTIONAL, -- No sig if clear key
OID= 1 3 14 3 2 11 RSA signature
OID= 1 3 14 3 2 29 RSA w/SHA1 signature
OID= 1 2 840 113549 1 1 5 sha1withRSAEncryption (from device)
kedSignature OCTET STRING OPTIONAL
The following is the PKCS #1 ASN.1 definition.

SubjectPublicKeyInfo ::= SEQUENCE {
algorithm AlgorithmIdentifier, -- (we intend to use RSA)
OID= 1 2 840 113549 1 1 1 RSA
subjectPublicKey BIT STRING
Where subjectPublicKey is:

RSAPublicKey ::= SEQUENCE {

modulus INTEGER, -- n
publicExponent INTEGER -- e
ASN.1 xAK/xSK Structure (Clear, Signed) - Example

ASN1 Data
0 4 572 : SEQUENCE {
: <30 82 02 73>
: <02 01 00>
7 11 290-: SEQUENCE { kedInfo
: <30 82 01 22> (may be 21)
11 13 13 : SEQUENCE { algorithm
: <30 0B>
13 15 9 : OID 1 2 840 113549 1 1 1
: <06 09 2A 86 48 86 F7 0D 01 01 01>
24 26 0 : NULL
: <05 00>
: }
26 30 271-: BIT STRING (0 unused bits) {subjectPublicKey
: <03 82 01 0F>
: <00>
31 35 266-: SEQUENCE { RSA key
: <30 82 01 0A> (may be 09)
35 39 257-: INTEGER modulus
: <02 82 01 01> (may be 00 [need to add 00 if msb=1])
: <[00]
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??>
296- 298- 3 : INTEGER publicExponent
: <02 03>
: <01 00 01>
: }
: }
: }
301- 303- 13 : SEQUENCE {kedSigAlgID
: <30 0D>
303- 305- 9 : OID 1 2 840 113549 1 1 5
: <06 09 2A 86 48 86 F7 0D 01 01 05>
314- 316- 0 : NULL
: <05 00>
: }
316- 318- 256 : OCTET STRING kedSignature
: <04 82 01 00>
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??

: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??>
: }
Signing ASN.1 Format

The following structure is used to specify or negotiate the HMAC.
SigningInfo ::= SEQUENCE {
hmacMethod MethodHMAC -- specify/select HMAC method
siSigAlgID AlgorithmIdentifier OPTIONAL,
OID= 1 3 14 3 2 11 RSA signature
OID= 1 2 840 113549 1 1 5 sha1withRSAEncryption
siSignature OCTET STRING OPTIONAL
MethodHMAC ::= CHOICE {

hmacSelection[0] EXPLICIT SelectHMAC-- used by EKM to select method
hmacEncrypted[1] EXPLICIT ProtHMAC-- eAK encrypted ClearHMAC
hmacClear[2] EXPLICIT ClearHMAC-- only used when no eAK
SelectHMAC ::= SEQUENCE {

hmacAlg AlgorithmIdentifier, -- Algorithm used for message hashing
OID= 1 3 14 3 2 18 SHA
OID= 1 3 14 3 2 26 SHA1
OID= 1 3 14 3 2 11 RSA signature (not recommended)
hmacLen INTEGER -- length in bytes of HASH
ProtHMAC ::= SEQUENCE {

hmacEncAlg AlgorithmIdentifier, -- eAK method (RSA)
OID= 1 2 840 113549 1 1 1 RSA
OID= 1 2 840 113549 1 1 7 RSAES-OAEP
hmacEncOCTET STRING -- eAK encrypted ClearHMAC
ClearHMAC ::= SEQUENCE {

OID= 1 3 14 3 2 18 SHA
OID= 1 3 14 3 2 26 SHA1
hmacLen INTEGER -- length in bytes of HASH
hmacKey OCTET STRING -- The HMAC key (seed data)
}

ASN.1 Signing Information (Clear, Unsigned) - Example
ASN1 Data
0 2 56 : SEQUENCE {
: <30 38>
: <02 01 00>
5 7 51 : [2] SEQUENCE { hmacClear
: <A2 33>
: <02 01 00>
10 12 9 : SEQUENCE {hmacAlg
: <30 09>
12 14 5 : OID 1 3 14 3 2 18
: <06 05 2B 0E 03 02 12>
19 21 0 : NULL
: <05 00>
: }
21 23 1 : INTEGER 32 hmacLen
: <05 01 20>
24 26 32 : OCTET STRING hmacKey
: <04 20>
: <?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??>
: }
: }
ASN.1 Signing Information (eAK Encrypted, dAK Signed) - Example

ASN1 Data
0 4 560: SEQUENCE {
: <30 82 02 30> version
4 6 1 : INTEGER 0
: <02 01 00>
7 11 278: [1] SEQUENCE { hmacEncrypted
: <A1 82 01 16>
: <02 01 00>
14 16 13 : SEQUENCE { hmacEncAlg
: <30 0D>
16 18 9 : OID 1 2 840 113549 1 1 1
: <06 09 2A 86 48 86 F7 0D 01 01 01>
27 29 0 : NULL
: <05 00>
: }
29 33 256: OCTET STRING hmacEnc (w/eAK)
: <04 82 01 00>
: <?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??>
: }

289 291 13 : SEQUENCE {siSigAlgID
: <30 0D>
291 293 9 : OID 1 2 840 113549 1 1 5
: <06 09 2A 86 48 86 F7 0D 01 01 05>
302 304 0 : NULL
: <05 00>
: }
304 308 256: OCTET STRING siSignature
: <04 82 01 00>
: <?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??>
: }
HMAC ASN.1 Format

The following is taken from PKCS #1.
ClearHMAC ::= SEQUENCE {
OID= 1 3 14 3 2 18 SHA
OID= 1 3 14 3 2 26 SHA1
DigestInfo ::= SEQUENCE {

digestAlgorithmDigestAlgorithmIdentifier,
OID= 1 3 14 3 2 18 SHA
OID= 1 3 14 3 2 26 SHA1
digest Digest
DigestAlgorithmIdentifier ::= AlgorithmIdentifier
Digest ::= OCTET STRING
ASN.1 HMAC Structure - Example

ASN1 Data
0 4 47 : SEQUENCE {
: <30 2F>
: <02 01 00>
5 7 9 : SEQUENCE {digestAlgorithm
: <30 09>
7 8 5 : OID 1 3 14 3 2 18
: <06 05 2B 0E 03 02 12>
13 15 0 : NULL

: <05 00>
: }
15 17 32 : OCTET STRING digest
: <04 20>
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??>
: }

Appendix B. Notices
Trademarks
The following terms are trademarks of International Business Machines
Corporation in the United States or other countries (or regions) or both:
AIX OS/400 pSeries

AS/400 pSeries RAA
eServer SP zSeries
IBM RS/6000 NetBackup (NBU)
iSeries TotalStorage
The following are U. S. trademarks of Hewlett-Packard Company, International

Business Machines Corporation, and Seagate Technology:
Linear Tape-Open
LTO
Ultrium
Intel and Itanium are registered trademarks of Intel Corporation in the United
States, or other countries (or regions), or both.
Linux is a register trademark of Linus Torvalds.
NetBackup is a trademark of Symantec
Microsoft®, Windows®, Windows NT, and the Windows logo are registered
trademarks of Microsoft Corporation in the United States, or other countries (or
regions), or both.
Other company, product, and service names may be trademarks or service marks
of others.

Glossary
This glossary defines the special terms, cartridge memory. See LTO cartridge memory.
abbreviations, and acronyms that are used in this
circuit board. A thin sheet on which chips and other
publication.
electronic components are placed. Computers consist of
one or more boards, often called cards or adapters.
Numbers and Symbols
cleaning cartridge. A tape cartridge that is used to
2:1 compression. The relationship between the clean the heads of a tape drive. Contrast with data
quantity of data that can be stored with compression as cartridge.
compared to the quantity of data that can be stored
without compression. In 2:1 compression, twice as command timeout. Following the issuance of a
much data can be stored with compression as can be command, a period of time during which it is
stored without compression. determined that there is a bad connection between the
server and the drive.
A compression. The process of eliminating gaps, empty

fields, redundancies, and unnecessary data to shorten
adapter. See adapter card. the length of records or blocks.
adapter card. A circuit board that adds function to a configure. To describe to a server the devices, optional
computer. features, and programs installed on the system.
AL_PA. See Arbitrated Loop Physical Address.

D
Arbitrated Loop Physical Address (AL_PA). An 8-bit
value that identifies a device in an arbitrated loop. All data. Any representations such as characters or analog
Fibre Channel ports communicate by using AL_PAs. quantities to which meaning is, or might be, assigned.
data cartridge. A tape cartridge that is dedicated to

B storing data. Contrast with cleaning cartridge.
backups. The short-term retention of records used for data compression. See compression.
restoring essential business and server files when vital
data has been lost because of program or server errors data transfer rate. The average number of bits,
or malfunctions. characters, or blocks per unit of time that pass between
corresponding equipment in a data transmission
bezel. The removable frame that fits over the front of system. The rate is expressed in bits, characters, or
the Ultrium Tape Drives. blocks per second, minute, or hour.
bit. The smallest unit of data in a computer. A bit device. Any hardware component or peripheral, such
(short for binary digit) has a single binary value (either as a tape drive or tape library, that can receive and
0 or 1). Computers store data and execute instructions send data.
in bit multiples called bytes. In most computer systems,
there are eight bits in a byte. device driver. A binary file that is installed on a host
system and enables the host system to access a device.
bus. See SCSI bus.
diagnostic. A software program that is designed to
byte. A string that consists of a certain number of bits recognize, locate, and explain faults in equipment or
(usually 8) which are treated as a unit and represent a errors in programs.
character. A byte is a fundamental unit of data.
diagnostic cartridge. A tape cartridge that enables the
detection and isolation of errors in programs and faults
C in equipment.
capacity. The amount of data that can be contained on drive. See IBM Ultrium Internal Tape Drive Models T200
storage media and expressed in bytes. and T200F or IBM TotalStorage LTO Ultrium 2 Tape Drive
Models T400 and T400F.
cartridge. See tape cartridge.

drive dump. The recording, at a particular instant, of file. A named set of records that are stored or
the contents of all or part of one storage device into processed as a unit.
another storage device, usually as a safeguard against
faults or errors, or in connection with debugging. filemark. Located on the magnetic tape within a tape
cartridge, a recorded element that typically marks the
drive head. The component that records an electrical organizational boundaries in a serial file structure (such
signal onto magnetic tape, or reads a signal from tape as directory boundaries) and that is requested to be
into an electrical signal. written or read by the server.
drive sense data. See SCSI drive sense data. firmware. Proprietary code that is usually delivered as
part of an operating system. Firmware is more efficient
dump. See drive dump. than software that is loaded from an alterable medium,
and is more adaptable to change than pure hardware
duplex. See duplex transmission. circuitry. An example of firmware is the Basic
Input/Output System (BIOS) in read-only memory
duplex adapter. A mechanism that allows a device to
(ROM) on a PC motherboard.
send and receive communication at the same time.
duplex transmission. Transmission in both directions, G

either one direction at a time (half-duplex) or both
directions simultaneously (full-duplex). Gb. See gigabit.
E GB. See gigabyte.
gigabit (Gb). 1 000 000 000 bits.

eject. To remove or force out from within.
gigabyte (GB). 1 000 000 000 bytes.
electronic mail. Correspondence in the form of
messages transmitted between user terminals over a
computer network. H
e-mail. See electronic mail. hard addressing. Pertaining to the Fibre Channel
drives (Models T200F and T400F), a method that
enclosure. A device, such as a desktop unit, tape identifies the drive’s LID and, consequently, its AL_PA
cartridge autoloader, or tape library, into which you can (the AL_PA enables the drive to communicate with
install an Ultrium Tape Drive. other devices).
error log. Maintained by an Ultrium Tape Drive, a list hardware. The physical equipment or devices that
that contains the ten most recent error codes. The codes form a computer.
identify errors that pertain to the drive.
head. See drive head.
F host. The controlling or highest-level system in a data
communication configuration. Synonymous with server.
Fibre Channel. A full-duplex, serial communications
technology that is capable of interconnecting Ultrium
Tape Drives and servers. Fibre Channel technology I
combines features of the input/output (I/O) and
networking interfaces. IBM Ultrium Internal Tape Drive Models T200 and
T200F. A data-storage device that controls the
Fibre Channel cable. The cable that connects a Fibre movement of the magnetic tape in an IBM LTO Ultrium
Channel tape drive to another device. The conductive Tape Cartridge. The drive houses the mechanism (drive
element within the cable is constructed of either copper head) that reads and writes data to the tape. Its native
wires or optical fibers. Generally, copper wires are used data capacity is 100 GB per cartridge and up to 200 GB
for short distances (up to 30 meters or 98 feet); optical at 2:1 compression. Its native data transfer rate is 15
fibers are used for longer distances. Fiber-optic cabling MB per second and 30 MB per second at 2:1
is referred to by mode or the frequencies of light waves compression.
that are carried by a particular cable type. Multimode
fiber cables are generally used for distances up to 500 IBM TotalStorage LTO Ultrium 2 Tape Drive Models
meters (1640 feet) and with short-wave (780 nanometer) T400 and T400F. A data-storage device that controls
laser light. Single-mode fiber cables are used for the movement of the magnetic tape in an IBM LTO
distances greater than 500 m (1640 feet) and with Ultrium Tape Cartridge. The drive houses the
long-wave (1300 nanometer) laser light. mechanism (drive head) that reads and writes data to
the tape. Its native data capacity is 200 GB per

cartridge and up to 400 GB at 2:1 compression. Its signaling uses two wires to drive one signal over
native data transfer rate is 35 MB per second and 70 copper wire. The use of wire pairs reduces electrical
MB per second at 2:1 compression. noise and crosstalk.
ID. Identifier. LTO. See Linear Tape-Open.
initiator. In SCSI terms, a SCSI device that requests an LTO cartridge memory (LTO-CM). Within each LTO
I/O process to be performed by another SCSI device (a Ultrium Data Cartridge, an embedded electronics and
target). In many cases, an initiator can also be a target. interface module that can store and retrieve a
cartridge’s historical usage and other information.
input/output (I/O). Data that is provided to a
computer or data that results from computer LTO-CM. See LTO cartridge memory.
processing.
LTO-DC. See LTO Data Compression.
install. (1) To set up for use or service. (2) The act of
adding a product, feature, or function to a server or LTO Data Compression (LTO-DC). A method that
device either by a singular change or by the addition of compresses a server’s data before the drive writes it to
multiple components or devices. tape. LTO-DC detects but does not recompress or test
record boundaries and file markers (which are encoded
Internet. The worldwide collection of interconnected as control symbols). It also allows switching between
networks that use the Internet suite of protocols and compression and no compression within the data
permit public access. stream, which prevents data from expanding when the
drive compresses random or encrypted data.
I/O. See input/output.
LVD. See Low Voltage Differential.
J
M
jumper. (1) A tiny connector that fits over a pair of
protruding pins in a connector. A jumper can be moved magnetic tape. A tape with a magnetizable surface
to change electrical connectors. When in place, the layer on which data can be stored by magnetic
jumper connects the pins electrically. (2) To place a recording.
jumper on a connector pin.
MB. See megabyte.
L media. The plural of medium.
Linear Tape-Open (LTO). A type of tape storage media capacity. See capacity.
technology developed by the IBM Corporation,
Hewlett-Packard, and Seagate. LTO technology is an medium. A physical material in or on which data may
“open format” technology, which means that its users be represented, such as magnetic tape.
have multiple sources of product and media. The
Medium Auxiliary Memory (MAM). A non-volatile
“open” nature of LTO technology enables compatibility
memory. MAM is used to store data that describes the
between different vendors’ offerings by ensuring that
media and its contents.
vendors comply with verification standards.
megabyte (MB). 1 000 000 bytes.
load. Following the insertion of a tape cartridge into
the tape load compartment, the act of positioning the micrometer. One millionth of a meter (.000001 m).
tape (performed by the tape drive) for reading or Synonymous with micron. Abbreviated as µm.
writing by the drive’s head.
micron. One millionth of a meter (.000001 m).
load and unload cycle. The act of inserting a cartridge Synonymous with micrometer. Abbreviated as µm.
into a tape drive, loading the tape to load point,
rewinding the tape into the cartridge, and ejecting the microsecond. One millionth of a second (.000001 s).
cartridge from the drive. Abbreviated as µs.
log sense data. See SCSI log sense data. millimeter (mm). One thousandth of a meter (.001 m).
Low Voltage Differential (LVD). A low-noise, millisecond (ms). One thousandth of a second (.001 s).
low-power, and low-amplitude electrical signaling
system that enables data communication between a mm. See millimeter.
supported server and the Ultrium Tape Drive. LVD
ms. See millisecond.
Glossary 311
Model T200. The version of the IBM Ultrium Internal operating system. The master computer control
Tape Drive that uses the SCSI interface, has a native program that translates the user’s commands and
storage capacity of 100 GB, and a native data transfer allows software application programs to interact with
rate of 15 MB per second. the computer’s hardware.
Model T200F. The version of the IBM Ultrium Internal

Tape Drive that uses the Fibre Channel interface, has a P
native storage capacity of 100 GB, and a native data
transfer rate of 15 MB per second. parity. The state of being even-numbered or
odd-numbered. A parity bit is a binary number that is
Model T400. The version of the IBM TotalStorage LTO added to a group of binary numbers to make the sum
Ultrium 2 Tape Drive that uses the SCSI interface, has a of that group always odd (odd parity) or even (even
native storage capacity of 200 GB, and a native data parity).
transfer rate of 35 MB per second.
parity error. A transmission error that occurs when the
Model T400F. The version of the IBM TotalStorage received data does not have the parity that is expected
LTO Ultrium 2 Tape Drive that uses the Fibre Channel by the receiving system. This usually occurs when the
interface, has a native storage capacity of 200 GB, and a sending and receiving systems have different parity
native data transfer rate of 35 MB per second. settings.
port. (1) A system or network access point for data

N entry or exit. (2) A connector on a device to which
cables for other devices such as display stations and
N/A. Not applicable. printers are attached. (3) The representation of a
physical connection to hardware. A port is sometimes
native storage capacity. The amount of data that can referred to as an adapter; however, there can be more
be stored without compression on a tape cartridge. than one port on an adapter.
native sustained data transfer rate. See data transfer power-off. To remove electrical power from a device.
rate.
power-on, powered-on. (1) To apply electrical power
network. A configuration of data processing devices to a device. (2) The state of a device when power has
and software that is connected for information been applied to it.
interchange.
protocol. The meanings of, and the sequencing rules
network server. In a local area network, a personal for, requests and responses that are used to manage a
computer that provides access to files for all of the network, transfer data, and synchronize the states of
workstations in the network. network components.
node. In Fibre Channel technology, a communicating
device. Q
ntutil. Created by IBM, a utility program for LTO quiesce. To put a device into a temporarily inactive or
devices that connect to Windows NT and Windows inhibited state, but not remove it from the server.
2000. ntutil provides problem determination for
hardware or connections, assists with device and
medium changer recognition, forces dumps, loads new R
firmware, sends and receives SCSI commands to and
from the hardware, and obtains SCSI sense data to use read. To acquire or interpret data from a storage
in resolving errors. device, from a data medium, or from another source.
reboot. To reinitialize the execution of a program by

O repeating the initial program load (IPL) operation.
offline. The operating condition that the Ultrium Tape record. The smallest distinct set of data bytes that is
Drives are in when the server’s applications cannot supplied from a server for processing and recording by
interact with it. a tape drive, and the smallest distinct set of data to be
read from tape, reprocessed, and made available to a
online. The operating condition that the Ultrium Tape server by a tape drive.
Drives are in when the server’s applications can
interact with it. record boundaries. The fixed limits of a record.
Open Systems. Computer systems whose standards

are not proprietary.

repeater. A device that regenerates signals to extend SCSI ID connector. Located at the rear of the Ultrium
the range of transmission between data stations or to Tape Drive, the connector that enables the drive’s SCSI
interconnect two branches. A repeater is a node of a address to be set. Addresses are determined by the
local area network. placement of jumpers on the pins.
reset. To return a device or circuit to a clear state. SCSI ID switch. Located on an enclosure that
contains a Ultrium Tape Drive, a mechanism that
RS-422 connector. Located at the rear of the Ultrium connects to the drive and allows you to change the
Tape Drive, the connector to which the internal RS-422 drive’s SCSI ID without using jumpers.
cable of an enclosure connects. The connection enables
serial devices to communicate with the drive. SCSI log sense data. In response to inquiry from the
server about the Ultrium Tape Drive’s error logs and
RS-422 interface. An electrical interface standard that counters, a packet of SCSI sense bytes which contains
is approved by the Electronic Industries Association that information and which is sent back to the server
(EIA) for connecting serial devices. The RS-422 by the drive. Log sense data is used to diagnose
standard, which supports higher data rates and greater problems, especially if the problems are intermittent.
immunity to electrical interference, is an alternative to
the older RS-232 interface, and uses individual second. One sixtieth of a minute.
differential signal pairs for data transmission.
Depending on data transmission rates, RS-422 can be selection timeout. Following the selection of an option
used at distances up to 1,275 m (4,000 ft). The RS-422 (for example, a data transfer), the period of time during
interface also supports multi-point connections. which it is determined that there is a bad connection
between the server and the drive.
S sense data. Data that describes an I/O error. Sense

data is presented to a server in response to a Sense I/O
s. See second. command.
SAN. See Storage Area Network. Serial Attached SCSI (SAS). A transport for
exchanging information between SCSI devices using a
SAN Data Gateway. A device that provides Fibre serial interconnect.
Channel attachment between Open Systems servers and
SCSI disk and tape storage systems. serial interface. An interface that sequentially or
consecutively executes two or more operations in a
SAS. Serial Attached SCSI single device, such as an arithmetic and logic operation.
SC. See subscription channel connector. server. A functional unit that provides services to one
or more clients over a network. Examples include a file
SCSI. See Small Computer Systems Interface.
server, a print server, or a mail server. The IBM Eserver
SCSI bus. (1) A collection of wires through which pSeries, IBM Eserver iSeries, HP, and Sun are servers.
data is transmitted from one part of a computer to Synonymous with host.
another. (2) A generic term that refers to the complete
short-wave cable. In Fibre Channel technology, a laser
set of signals that define the activity of the Small
cable that uses a wavelength of 780 nanometers and is
Computer Systems Interface (SCSI).
only compatible with multimode fiber.
SCSI connector. Located at the rear of the Ultrium
Small Computer Systems Interface (SCSI). A
Tape Drive, the connector that facilitates commands to
standard used by computer manufacturers for attaching
and from the server, and to which the internal SCSI
peripheral devices (such as tape drives, hard disks,
cable of an enclosure connects.
CD-ROM players, printers, and scanners) to computers
SCSI device. Anything that can connect into the SCSI (servers). Pronounced ″scuzzy.″ Variations of the SCSI
bus and actively participate in bus activity. interface provide for faster data transmission rates than
standard serial and parallel ports (up to 160 megabytes
SCSI drive sense data. In response to inquiry from per second). The variations include:
the server about an error condition, a packet of SCSI v Fast/Wide SCSI: Uses a 16-bit bus, and supports data
sense bytes that contains information about the error rates of up to 20 MBps.
and that is sent back to the server by the drive. v SCSI-1: Uses an 8-bit bus, and supports data rates of
4 MBps.
SCSI ID. The unique address (from 1 to 15) that you
assign to an Ultrium Tape Drive that uses a SCSI v SCSI-2: Same as SCSI-1, but uses a 50-pin connector
interface. instead of a 25-pin connector, and supports multiple
devices.
Glossary 313
v Ultra SCSI: Uses an 8- or 16-bit bus, and supports tapeutil. Created by IBM, a utility program for LTO
data rates of 20 or 40 MBps. devices that connect to all supported servers except
v Ultra2 SCSI: Uses an 8- or 16-bit bus and supports Windows NT and Windows 2000. tapeutil provides
data rates of 40 or 80 MBps. service aids for tape subsystems, offers a menu-driven
tool for exercising or testing IBM tape and medium
v Ultra3 SCSI: Uses a 16-bit bus and supports data
changer devices, and includes a command-line interface
rates of 80 or 160 MBps.
that is convenient for use in shell scripts.
v Ultra160 SCSI: Uses a 16-bit bus and supports data
rates of 160 MBps. terminate. To prevent unwanted electrical signal
reflections by applying a device (known as a
soft addressing. Pertaining to the Fibre Channel drive, terminator) that absorbs the energy from the
a method that enables the drive to dynamically transmission line.
arbitrate its AL_PA with other Fibre Channel devices on
the loop. The AL_PA enables the drive to communicate topology. In communications, the physical or logical
with other devices. arrangement of nodes in a network, especially the
relationships among nodes and the links between them.
software. Programs, procedures, rules, and any
associated documentation pertaining to the operation of TotalStorage LTO Ultrium Tape Drive Models T400
a computer system. and T400F. See IBM TotalStorage LTO Ultrium 2 Tape
Drive Models T400 and T400F.
Storage Area Network (SAN). A high-speed
subnetwork of shared storage devices. A SAN’s transfer rate. See data transfer rate.
architecture makes all storage devices available to all
servers on a LAN or WAN. As more storage devices
are added to a SAN, they too will be accessible from
U
any server in the larger network. Because stored data
Ultrium Tape Drive. See IBM Ultrium Internal Tape
does not reside directly on any of a network’s servers,
Drive Models T200 and T200F or IBM TotalStorage LTO
server power is used for business applications, and
Ultrium 2 Tape Drive Models T400 and T400F.
network capacity is released to the end user.
Ultrium 2 Tape Drive. See IBM TotalStorage LTO
subscription channel connector (SC). A push-pull
Ultrium 2 Tape Drive.
type of optical connector that features high density, low
loss, low backreflection, and low cost. uniform resource locator (URL). The address of an
item on the World Wide Web. It includes the protocol
switch. A network infrastructure component to which
followed by the fully qualified domain name
multiple nodes attach. Unlike hubs, switches typically
(sometimes called the host name) and the request. The
have the ability to switch node connections from one to
web server typically maps the request portion of the
another. A typical switch can facilitate several
URL to a path and file name. For example, if the URL
simultaneous bandwidth transmissions between
is http://www.networking.ibm.com/nsg/nsgmain.htm,
different pairs of nodes.
the protocol is http; the fully qualified domain name is
www.networking.ibm.com; and the request is
T /nsg/nsgmain.htm.
TapeAlert. A patented technology and ANSI standard unload. The act (performed by the drive) of
that defines conditions and problems that are unthreading tape from the drive’s internal tape path
experienced by tape drives. and returning it (with the leader block) to the tape
cartridge.
TapeAlert flags. Status and error messages that are
generated by the TapeAlert utility and display on the URL. See uniform resource locator.
server’s console.
utility. See utility program.
tape cartridge. A removable storage case that houses
belt-driven magnetic tape that is wound on a supply universal time (UT). The time at longitude zero,
reel and a takeup reel. colloquially known as Greenwich Mean Time. See
http://aa.usno.navy.mil/faq/docs/UT.html.
tape drive. See IBM Ultrium Internal Tape Drive Models
T200 and T200F or IBM TotalStorage LTO Ultrium 2 Tape utility program. A computer program that supports
Drive Models T400 and T400F. computer processes. For example, a diagnostic
program, a trace program, or a sort program.
tape path. Within a tape drive, the channel in which
the media moves.

V
vital product data (VPD). Information about a
product. Among other details, the VPD may include a
model number, serial number, part number, or level of
firmware.
W
web. See World Wide Web.
World Wide Name. A unique, 8-byte identifier that is

assigned by IBM Manufacturing to each Ultrium Tape
Drive and used to identify a drive.
World Wide Web. A network of servers that contain

programs and files. Many of the files contain hypertext
links to other documents that are available through the
network.
write. To make a permanent or transient recording of

data in a storage device or on a data medium.
write protected. Applicable to a tape cartridge, the

condition that exists when some logical or physical
mechanism prevents a device from writing on the tape
in that cartridge.
Glossary 315
Index
Special characters D Firmware Update Behavior (Byte 7) 141
Force Dump diagnostic 237
(Not LTO1), REPORT SUPPORTED TASK Data Cartridge Emulates WORM Format for the Attribute List Service
MANAGEMENT FUNCTIONS 192 (01h) 138 Action 160
Data Compression Log 80 Format for the Attribute Values Service
Data Compression Mode Page 118 Action 159
A Data Encryption Algorithm Descriptor -
Standard Encryption 209
Format for the Partition List Service
abort condition, handling 282 Action 162
data transfer rate 2 Format for the Volume List Service
AS/400
Deferred Sense data 261 Action 161
device driver support 4
description of tape drives 1 FUNCTIONS (Not LTO1), REPORT
requirements for attaching tape drive
Device Clocks 265 SUPPORTED TASK
to 3
device drivers, supported 4 MANAGEMENT 192
ASN.1 [DER] Encoding
Device Hardware Encryption (Ultrium
Implementation 297
4) 279
ASN.1 HMAC Structure - Example 304
Device Identification Page 39
ASN.1 Information 297
disconnect strategy 286 H
ASN.1 Signing Information (clear, Hewlett-Packard
Disconnect/Reconnect Page 115
unsigned) - Example 303 device driver support 4
Download Process 295
ASN.1 Signing Information (eAK requirements for attaching tape drive
Download, Firmware 293
encrypted, dAK signed) - Example 303 to 3
Identifying Level Hardware of
ASN.1 Standalone SEDK (SEDK only, high voltage differential (HVD) SCSI
Drive 293
clear, signed) - Example 298 interface 1, 2
Drive Component Revision Level
ASN.1 Standalone SEDK (SEDK only, HMAC ASN.1 Format 304
Pages 44
dSK encrypted, signed) - Example 299 Host Port 0 Interface Errors log page 97
Drive Control Statistics log page 105
ASN.1 xAK/xSK Structure (clear, signed) Host Port 1 Interface Errors 99
Drive Control Verification log page 98
- Example 301 HP
attaching tape drive to server 3 device driver support 4
Attachment Features 281
E requirements for attaching tape drive
to 3
Encryption 297 HVD interface 1, 2
B Engineering Use log page 104
ERASE command 17, 20
backward error page 85
Error Log Buffer (06h) 179
Blocks/Bytes Transferred log page 95
Error Log Buffer (06h)r 177 I
burst data transfer rate 2 Implementation Considerations 265
error sense information 261
Bus Parity Errors (when using Information Exceptions Mode Page 130
errors
Information unit transfers) 286 INQUIRY command 17, 21
bus parity 285
bus parity errors, handling 285 Intel-compatible servers
persistent 262
C F to 3
capacity 2, 5 interface
Features of the Serial Attached SCSI HVD 1, 2
Clean Behavior (Byte 3) 137
(SAS) Interface 291 LVD 1, 2
Clear FLASH Dump (EEPROM) 241
Fence Behavior (Byte 2) 135 interface attachment 281
Command Behavior When WORM
Fencing Behavior 262 InterNational Committee for Information
Medium Has Been Tampered With 267
Fibre Channel attachment Technology Standards (INCITS) 17
Command Descriptor Block (CDB) 17
abort condition, handling 281, 282 iSeries server
command(s)
features of interface 291 device driver support 4
ERASE 20
LUN behaviors 281 requirements for attaching tape drive
INQUIRY 21
multi-initiator support 281, 284 to 3
LOAD/UNLOAD 45
power-on procedure 281
LOCATE 47
requirements 3
LOG SELECT 49
reset strategy 281
LOG SENSE 50
status codes 281, 284 K
supported SCSI 17 KAD 00h - Unauthenticated KAD 219
Fibre Channel Logical Unit Control
compatible tape cartridges 5 KAD 01h - DKi (Data Key
Page 123
compressed capacity 2 Identifier) 220
Firmware Download 293
Conditions for Writing 267 KAD 02h - Nonce 221
Identifying Level Hardware
Control Extension Mode Page 117 KAD 03h - Metadata (Raw) 222
Firmware image is intended 295
Current Sense data 261 Key-Associated Data (KAD)
Identifying Level Hardware of
Drive 293 Descriptors 214, 219

L Mode Page 2Fh: Behavior Configuration
Mode Page 134
READ POSITION 180
READ POSITION command 17
LC-duplex 2 Mode Page 3Dh: LEOT Mode Page (3Dh) Read Thermal Sensor 245
Linux (LTO 1 Only) 143 Read-Write Error Recovery Page 113
device driver support 4 Mode Parameter Header 110 read/write capability 5
requirements for attaching tape drive MODE SELECT (10) command 17, 106 RECEIVE DIAGNOSTIC RESULTS 183
to 3 MODE SELECT (6) command 17, 106 RECEIVE DIAGNOSTIC RESULTS
LOAD/UNLOAD command 17 MODE SENSE (10) command 17, 107 command 17
LOAD/UNLOAD Command 45 MODE SENSE (6) command 17, 107 RELEASE UNIT (10) command 18, 184
LOCATE command 17, 47 multi-initiator support 281, 284 RELEASE UNIT (6) command 17, 184
Log Page 11h: DT Device Status Log Remove Traps diagnostic 243
Page 62 REPORT DENSITY SUPPORT 185
Log Page 37h 85
Log Page format 51 N REPORT DENSITY SUPPORT
command 18
log page(s) native storage capacity 2
REPORT LUNs 190
Blocks/Bytes Transferred 95 native sustained data transfer rate 2
REPORT LUNs command 18
Data Compression 80 Negotiated Settings mode subpage 127
REPORT SUPPORTED TASK
Drive Control Statistics 105 Non-Medium Errors log page 57
MANAGEMENT FUNCTIONS (Not
Drive Control Verification 98 Notices 307
LTO1) 192
Engineering Use 104 REPORT TIMESTAMP (Not LOT 1) 193
Host Port 0 Interface Errors 97 Report Transfer Capabilities mode
Host Port 1 Interface Errors 99 O subpage 129
Non-Medium Errors 57 operating systems, supported 3 Reporting Behavior (Byte 6) 140
Read Error Counters 56 REQUEST SENSE 194
Read Forward Errors 83 REQUEST SENSE command 18
Sequential Access Device 58
Subsystem Statistics 101 P requirements for attaching tape drive to
server 3
supported 52 persistent errors 262
RESERVE UNIT (10) command 18, 199
Tape Capacity 79 PERSISTENT RESERVE IN 144
RESERVE UNIT (6) command 18, 199
Tape Usage 77 PERSISTENT RESERVE IN command 17
Reset Drive diagnostic 245
TapeAlert 59 PERSISTENT RESERVE OUT 149
reset strategy 281
Write Error Counters 55 PERSISTENT RESERVE OUT
REWIND 200
Write Errors 81 command 17
REWIND command 18
LOG SELECT command 17, 49 Plaintext Key Format (00h) 210
RS-422 interface 2
LOG SENSE command 17, 50 Port Identifier 236
RS/6000
low voltage differential (LVD) SCSI Post A self test diagnostic 231
interface 1, 2 Post B performance diagnostic 232
LTO1), REPORT SUPPORTED TASK Post B Performance Diagnostic 232
to 3
MANAGEMENT FUNCTIONS Post C media test diagnostic 234
RS/6000 SP
(Not 192 Post D head test diagnostic 235
LUN identification 285 power-on procedure 281
LVD interface 1, 2 PREVENT/ALLOW MEDIUM
to 3
REMOVAL 153
PREVENT/ALLOW MEDIUM
M REMOVAL command 17
Primary Port Wrap Test Diagnostic 236 S
MAM Protocol Specific Port Control Page short SAN connection 2
description 163 format (Ultra 320 attached drives SC-duplex 2
format 164 only) 126 SCSI attachment
MANAGEMENT FUNCTIONS (Not pSeries server abort condition, handling 281, 282
LTO1), REPORT SUPPORTED device driver support 4 features of interface 285
TASK 192 requirements for attaching tape drive LUN behaviors 281
Medium Auxiliary Memory (MAM) to 3 multi-initiator support 281, 284
description 163 power-on procedure 281
format 164 requirements 3
message in 289
message out 287 R reset strategy 281
status codes 281, 284
messages 287, 289 READ 154
SECURITY PROTOCOL IN (SPIN)
MIM data structure 227 READ ATTRIBUTE 156
A2h 201
Mode Block Descriptor 112 READ ATTRIBUTE command 17
Security Protocol In Pages (00h - Security
Mode data format 109 Read Backwards Errors 85
Protocol Information) 202
Mode Page 19h: Protocol Specific Port READ BLOCK LIMITS 172
Security Protocol In Pages (20h - Tape
Control Page 124 READ BLOCK LIMITS command 17
Data Encryption) 205
Mode Page 1Ch: Information Exceptions READ BUFFER 173
SECURITY PROTOCOL OUT (SPOUT)
Mode Page 130 READ BUFFER command 17
B5h 215
Mode Page 1Dh: Medium Configuration READ command 17
Security Protocol Out Pages (20h - Tape
Mode Page 131 Read Error Counters log page 56
Data Encryption security protocol) 216
Read Forward Errors log page 83
SEDK ASN.1 Format 297

Self-Test diagnostic 230 supported device drivers 4 Write Dump to Flash (EEPROM) 240
SEND DIAGNOSTIC 223 supported SAN components 3 Write Error Counters log page 55
SEND DIAGNOSTIC command 18, 223 supported servers and operating Write Errors log page 81
Sense data systems 3 WRITE FILE MARKS command 18, 260
description 196 supported tape cartridges 5
types of and management 261 SUPPORTED TASK MANAGEMENT
Sense Data Behavior (Byte 5) 139
Sense Data Format 194
FUNCTIONS (Not LTO1),
REPORT 192
X
xAK/xSK ASN.1 Format 300
Sense Key 0 (No Sense) 269
Sense Key 1 (Recovered Error) 270
Sense Key 2 (Not Ready) 271
Sense Key 3 (Medium Error) 271
T
Tape Capacity Log 79
Sense Key 4 (Hardware or Firmware
tape cartridge
Error) 273
capacity 5
Sense Key 5 (Illegal Request) 273
compatible cartridges 5
Sense Key 6 (Unit Attention) 275
tape drives
Sense Key 7 (Data Protect) 275
description 1
Sense Key 8 (Blank Check) 276
Tape Usage Log 77
Sense Key B (Aborted Command) 277
TapeAlert log page 59
Sense Key D (Volume Overflow) 277
TASK MANAGEMENT FUNCTIONS
Sense keys and additional Sense 269
(Not LTO1), REPORT
Sequential Access Device Configuration
SUPPORTED 192
Page 120
TEST UNIT READY command 18, 254
Sequential Access Device log page 58
Trademarks 307
servers, supported 3
transfer rate 2
SET CAPACITY command 18, 248
Tru64
SET TIMESTAMP (Not LTO 1) 250
Set Traps diagnostic 242
Types of Interface Attachments 281
Signing ASN.1 Format 302
SIM data structure 224
SPACE (6) command 252
SPACE command 18 U
SPI Information Unit (Ultra 320 Attached Ultra 160 LVD SCSI interface 2
drives only) 285 Unit Attention Conditions 261
SPIN (00h:0000h) - Supported Security Unit Attention Sense data 261
Protocols List 203 Unit Serial Number Page 38
SPIN (00h:0001h) - Certificate Data 204 Unload On Error Behavior (Byte 8) 142
SPIN (20h:0000h) - Tape Data Encryption
In Support Pages page 206
SPIN (20h:0001h) - Tape Data Encryption
Out Support Pages page 207
V
Vendor-Specific Control Mode Page 133
SPIN (20h:0010h) - Data Encryption
Vendor-specific Medium Type
Capabilities page 208
Attributes 158
SPIN (20h:0011h) - Supported Key
Vendor-Specific Speed Matching Control
Formats page 210
Mode Page 133
VERIFY command 18, 255
Management Capabilities 211
viewing latest supported attachments 4
Vital Product Data Page 27
Status page 212
SPIN (20h:0021h) - Next Block Encryption
Status page 213
SPOUT (20h:0010h) - Set Data W
Encryption 216 Windows 2000
Standard Inquiry Data 22 device driver support 4
status codes 281, 284 requirements for attaching tape drive
Storage Area Network (SAN) to 3
connection 2 Windows NT
storage capacity 2 device driver support 4
Subsystem Statistics log page 101 requirements for attaching tape drive
Sun Microsystems to 3
device driver support 4 World Wide Name Buffer 178
requirements for attaching tape drive WORM Behavior (Byte 4) 138
to 3 WORM BEHAVIORS 267
supported WRITE ATTRIBUTE command 18, 257
log pages 52 WRITE BUFFER command 18, 258
SCSI commands 17 WRITE command 18, 256
Vital Product Data page 27 Write Dump to Cartridge diagnostic 238
Index 319

Printed in USA
GA32-0450-09

Ga32 0450 09

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ga32 0450 09

Uploaded by

Copyright:

Available Formats

IBM TotalStorage LTO Ultrium

Tape Drive

Tenth Edition (May 2008)

Second Edition, April 2004

Third Edition, December 2004

© Copyright IBM Corp. 2002, 2008 iii

Fourth Edition, March 2005

Fifth Edition, May 2006

Sixth Edition, December 2006

Seventh Edition, June 2007

iv IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Eighth Edition, July 2007

Ninth Edition, August/September 2007

Read This First v

Tenth Edition, May 2008

vi IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

© Copyright IBM Corp. 2002, 2008 vii

viii IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Appendix A. ASN.1 Information (Used

© Copyright IBM Corp. 2002, 2008 xi

xii IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

xiv IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Portions of this manual were adapted from documentation provided by the

xvi IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

© Copyright IBM Corp. 2002, 2008 1

2 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Primary Interface Attachment

IBM provides PostScript- and PDF-formatted versions of its documentation in the

4 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

LTO Data Capacity Data Capacity Supported by

6 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

© Copyright IBM Corp. 2002, 2008 7

Differences in Command Timeout Values

It is strongly recommended that device drivers or host software implement device

8 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Chapter 2. Drive Generation Differences 9

10 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Chapter 2. Drive Generation Differences 11

Table 6. Command Timeout Values (Ultrium 3 Half-Height and Ultrium 4) - Alphabetical

12 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Chapter 2. Drive Generation Differences 13

14 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Data Changes Between Generations

Standard Inquiry Data

Product Identification information returned is shown in Table 13 on page 26.

For more information, see “Standard Inquiry Data” on page 22.

REPORT DENSITY SUPPORT Command

For more information, see “REPORT DENSITY SUPPORT” on page 185.

Chapter 2. Drive Generation Differences 15

READ POSITION Command

Request Sense Changes

Behavior Changes Between Generations

Cartridge Eject for Errors

Microcode Detection of Errors

16 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

© Copyright IBM Corp. 2002, 2008 17

18 IBM TotalStorage LTO Ultrium Tape Drive: SCSI Reference

Note 2: Applicable Conditions are as follows:

In all other cases, the command will be processed as defined in SPC-2.

In all other cases, the command will be processed as defined in SPC-2.

Chapter 3. Command Support 19

EOD is written at the current position, which marks it as end of data.